You are on page 1of 736

Multi-hazard Loss Estimation Methodology

Earthquake Model


Hazus

MH 2.0

Technical Manual






Developed by:
Department of Homeland Security
Federal Emergency Management Agency
Mitigation Division
Washington, D.C.
ThismanualisavailableontheFEMAwebsiteatwww.fema.gov/plan/prevent/hazus.
ToordertheHazussoftwarevisitwww.msc.fema.govandgototheProductCatalogtoplace
yourorder
HazusisatrademarkoftheFederalEmergencyManagementAgency.

ACKNOWLEDGMENTS
HAZUSMH,HAZUSMHMR1,HAZUSMHMR2,HAZUSMHMR3,HAZUSMHMR4andHAZUS
MHMR5
EarthquakeModelMethodologyDevelopment
EarthquakeCommittee
Chairman,WilliamHolmes,Rutherford&Chekene,SanFrancisco,California
RogerBorcherdt,U.S.GeologicalSurvey,MenloPark,California
DavidBrookshire,UniversityofNewMexico,Albuquerque,NewMexico
RichardEisner,CaliforniaOfficeofEmergencyServices,OaklandCalifornia
RobertOlson,RobertOlson&Associates,Inc.,Sacramento,California
MichaelO'Rourke,RensselaerPolytechnicInstitute,Troy,NewYork
HenryLagorio,UniversityofCaliforniaatBerkeley,Berkeley,California
RobertReitherman,ConsortiumofUniversitiesforResearchinEarthquakeEngineering,
Richmond,California
RobertWhitman,MassachusettsInstituteofTechnology,Cambridge,Massachusetts,
ChairmanEmeritus
BuildingDamageSubcommittee
WilliamHolmes,Rutherford&Chekene,SanFrancisco,California
RobertWhitman,MassachusettsInstituteofTechnology,Cambridge,Massachusetts
ShakeBetaSubcommittee
Chairman,WilliamHolmes,Rutherford&Chekene,SanFrancisco,California
RobertWhitman,MassachusettsInstituteofTechnology,Cambridge,Massachusetts
RogerBorcherdt,U.S.GeologicalSurvey,MenloPark,California
RichardEisner,FAIA,CaliforniaOfficeofEmergencyServices,OaklandCalifornia
Error!Referencesourcenotfound.MichaelORourke,RennselaerPolytechnicInstitute,Troy,
NewYork
CasualtiesSubcommittee
Chairman,RobertReitherman,ConsortiumofUniversitiesforResearchinEarthquake
Engineering,Richmond,California
RichardEisner,CaliforniaOfficeofEmergencyServices,OaklandCalifornia
WilliamHolmes,Rutherford&Chekene,SanFrancisco,California
RobertOlson,RobertOlson&Associates,Inc.,Sacramento,California
HenryLagorio,UniversityofCaliforniaatBerkeley,Berkeley,California
PBS&J,Atlanta,Georgia
JawharBouabid,ProgramManager;ScottLawson
Kircher&Associates,PaloAlto,California
CharlesKircher
SanJoseStateUniversityFoundation,SanJose,California
ThaliaAnagnos
SpecialthankstotheEnvironmentalProtectionAgencyforitsassistanceinintegratingALOHA
intotheEarthquakeModel.
SpecialthankstotheNationalOceanicandAtmosphericAdministrationformakingFloodWAV
andFloodVIEWavailableanditsassistanceinintegratingthemintotheEarthquakeModel.
EarthquakModelValidation
ComartinReis,Stockton,California
CraigComartin,EvanReis

EarthquakeModelSoftwareDevelopment
SoftwareCommittee
Chairman,DickBilden,Consultant,Reston,Virginia
CoChairman,MikeHaecker,Consultant,Austin,Texas
DanCotter,Terrapoint,TheWoodlands,Texas
GerryKey,ComputerSciencesCorporation,SanDiego,California
TracyLenocker,LenockerandAssociates,Inc.,Orange,California
KenLewis,KVLandAssociates,Inc.,Scottsdale,Arizona
FrankOpporto,DHS,EP&RDirectorate(FEMA),InformationServicesTechnologyDivision,
Washington,D.C.
DirkVandervoort,POWEREngineers,Inc.,BoiseIdaho
LeslieWeinerLeandro,DHS,EP&RDirectorate(FEMA),InformationServicesTechnology
Division,Washington,D.C.
PBS&J,Atlanta,Georgia
MouradBouhafs,SeniorGroupManager;SandeepMehndiratta,NabilBouhafs,Eduardo
Escalona,RodErickson,MichellePalmer,PushpendraJohari,FouedBouhafs
BetaTestSubcommitteeHAZUSMH
DarrylDavis,CorpsofEngineersHydrologicEngineeringCenter,Davis,California
NeilGrigg,ColoradoStateUniversity,FortCollins,Colorado
CharlesKircher,Kircher&Associates,PaloAlto,California
TracyLenocker,LenockerandAssociates,Inc.,Orange,California
KennethLewis,KVLandAssociates,Inc.,Scottsdale,Arizona
MasoudZadeh,Consultant,SanJose,California

BetaTestCommunitiesHAZUSMH
DivisionofEmergencyManagement,Tallahassee,Florida;WashingtonStateEmergency
Management,CampMurray,Washington;WhatcomCountyPublicWorks,Bellingham,
Washington;JohnsonCounty,Olathe,Kansas;MecklenburgCountyStormwaterServices,
Charolotte,NorthCarolina;LouisianaStateUniversity,BatonRouge,Louisiana;Charleson
CountyBuildingServices,NorthCharleston,SouthCarolina
BetaTestSubcommitteeHAZUSMHMR1
DouglasBausch,DepartmentofHomelandSecurity,EmergencyPreparednessandResponse
Directorate(FEMA),Washington,D.C.
RichardEisner,Governor'sOfficeofEmergencyServices,Oakland,California
JohnKnight,SouthCarolinaEmergencyManagementDivision,Columbia,SouthCarolina
KevinMickey,ThePolisCenter,Indianapolis,Indiana
MarkNeimeister,DelawareGeologicalSurvey,Newark,Delaware
LynnSeirup,NewYorkCityOfficeofEmergencyManagement,NewYork,NewYork
BetaTestSubcommitteeHAZUSMHMR2
DouglasBausch,DepartmentofHomelandSecurity,FederalEmergencyManagement
Agency,Washington,D.C.
JohnKnight,SouthCarolinaEmergencyManagementDivision,Columbia,SouthCarolina
KevinMickey,ThePolisCenter,Indianapolis,Indiana
JoeRachel,DepartmentofHomelandSecurity,FederalEmergencyManagementAgency,
Washington,D.C.
KenWallace,DepartmentofHomelandSecurity,FederalEmergencyManagementAgency,
Washington,D.C.
BryanSiltanen,AdvancedSystemsDevelopment,Inc.,Arlington,VA
BetaTestSubcommitteeHAZUSMHMR3
DouglasBausch,DepartmentofHomelandSecurity,FederalEmergencyManagement
Agency,Washington,D.C.
JohnBuechler,ThePolisCenter,Indianapolis,IN
SusanCutter,HazardsandVulnerabilityResearchInstitute,UniversityofSouthCarolina,
Columbia,SC
AijuDing,CivilTechEngineering,Inc.,Cypress,TXs
CraigEissler,GeoTechVisualPowerTecTexasGeographicSociety,Austin,TX
MelanieGall,HazardsandVulnerabilityResearchInstitute,UniversityofSouthCarolina,
Columbia,SC
ShaneHubbard,ThePolisCenter,Indianapolis,IN
DavidMaltby,GeoTechVisualPowerTech/FirstAmericanFloodDataServices,Austin,TX
KevinMickey,ThePolisCenter,Indianapolis,IN
DavidNiel,APPTISInc.,McLean,VA
JackSchmitz,ThePolisCenter,Indianapolis,IN
BryanSiltanen,APPTISInc.,McLean,VA
HilaryStephens,Greenhorne&O'Mara,Inc.,Laurel,MD
EricTate,HazardsandVulnerabilityResearchInstitute,UniversityofSouthCarolina,
Columbia,SC
BetaTestingHAZUSMHMR4
KevinMickey,ThePolisCenter,Indianapolis,IN
JackSchmitz,ThePolisCenter,Indianapolis,IN
AdamCampbell,ThePolisCenter,Indianapolis,IN
ShaneHubbard,UniversityofIowa,IowaCity,IA
BetaTestingHAZUSMHMR5
KevinMickey,ThePolisCenter,Indianapolis,IN
JackSchmitz,ThePolisCenter,Indianapolis,IN
AdamCampbell,ZimmermanAssociates,Inc.,Hanover,MD
ShaneHubbard,UniversityofIowa,IowaCity,IA

HAZUSShellDevelopment
PBS&J,Atlanta,Georgia
MouradBouhafs,SeniorGroupManager;SandeepMehndiratta,NabilBouhafs,Eduardo
Escalona,RodErickson,MichellePalmer,PushpendraJohari,FouedBouhafs
SpecialthankstoESRIforitsassistanceincoordinatingArcGISwithHAZUSMH.

ProjectSponsorshipandOversight
DepartmentofHomelandSecurity,FederalEmergencyManagementAgency,Mitigation
Division,Washington,D.C.
FrederickSharrocks,SectionChief,Assessment&PlanforRisk;CliffOliver,Chief,Risk
AssessmentBranch;EdwardLaatsch,Chief,BuildingScienceandTechnology;EricBerman,
ProjectOfficer;ClaireDrury,ProjectOfficer;PaulTertell,MichaelMahoney,StuartNishenko,
ScottMcAfee,PaulBryant
FEMATechnicalMonitors
DouglasBausch,FEMARegion8;MarkCrowell,PhysicalScientist;JohnIngargiola,Douglas
Bellomo,AllysonLichtenfels,DivisionalCoordination

TABLEOFCONTENTS

PageNo.
1. IntroductiontoFEMALossEstimationMethodology
1.1 Background.....................................................................................................11
1.2 TechnicalManualScope..................................................................................11
1.3 TechnicalManualOrganization......................................................................12
2. OverallApproachandFrameworkofMethodology
2.1 VisionStatement.............................................................................................21
2.2 ProjectObjectives...........................................................................................24
2.3 DescriptionofLossEstimationMethodology.................................................25
3. InventoryData:CollectionandClassification
3.1 Introduction....................................................................................................31
3.2 DirectDamageDataBuildingsandFacilities.................................................31
3.2.1 GeneralBuildingStock.......................................................................31
3.2.1.1Classification..........................................................................32
3.2.1.2SpecificOccupancytoModelBuildingType
Mapping..............................................................................35
3.2.2 EssentialFacilities............................................................................330
3.2.2.1Classification........................................................................330
3.2.2.2OccupancytoModelBuildingTypeMapping......................331
3.2.3 HighPotentialLossFacilities............................................................331
3.2.3.1Classification........................................................................332
3.3 DirectDamageDataTransportationSystems.............................................332
3.3.1 HighwaySystems.............................................................................333
3.3.2 Railways...........................................................................................334
3.3.3 LightRail...........................................................................................335
3.3.4 Bus....................................................................................................336
3.3.5 PortsandHarbors............................................................................336
3.3.6 Ferry ................................................................................................337
3.3.7 Airports............................................................................................337
3.4 DirectDamageDataLifelineUtilitySystems...............................................338
3.4.1 PotableWaterSupply......................................................................338
3.4.2 WasteWater....................................................................................339
3.4.3 OilSystems.......................................................................................340
3.4.4 NaturalGasSystems........................................................................341
3.4.5 ElectricPower..................................................................................342
3.4.6 Communication................................................................................342
3.5 HazardousMaterialsFacility.........................................................................343
3.6 DirectEconomicandSocialLoss...................................................................343
3.6.1 DemographicsData..........................................................................343
3.6.2 DefaultOccupancyClassSquareFootInventory.............................346
3.7 IndirectEconomicData.................................................................................346
3.8 References.....................................................................................................348
Appendix3A ................................................................................................349
Appendix3B ................................................................................................367
Appendix3C ................................................................................................375
4. PotentialEarthScienceHazards(PESH)
4.1 GroundMotion...............................................................................................41
4.1.1 Introduction.......................................................................................41
4.1.1.1FormofGroundMotion/SiteEffects...................................42
4.1.1.2InputRequirementsandOutputInformation.......................43
4.1.2 DescriptionofMethods.....................................................................44
4.1.2.1 BasisforGroundShaking......................................................45
4.1.2.2 StandardShapeofResponseSpectra...................................47
4.1.2.3 AttenuationofGroundShaking..........................................410
4.1.2.4 SourcetoSiteDistanceMeasuresforAttenuationFunctions413
4.1.2.5 AmplificationofGroundShakingLocalSite
Conditions...........................................................................415
4.1.3 GuidanceforExpertGeneratedGroundMotionEstimation...........418
4.2 GroundFailure..............................................................................................421
4.2.1 Introduction.....................................................................................421
4.2.1.1Scope....................................................................................421
4.2.1.2InputRequirementsandOutputInformation.....................421
4.2.2 DescriptionofMethods...................................................................421
4.2.2.1Liquefaction.........................................................................421
4.2.2.1.1Background.........................................................422
4.2.2.1.2LiquefactionSusceptibility.................................422
4.2.2.1.3ProbabilityofLiquefaction.................................423
4.2.2.1.4PermanentGroundDisplacements....................428
4.2.2.2Landslide..............................................................................431
4.2.2.2.1Background.........................................................431
4.2.2.2.2LandslideSusceptibility......................................431
4.2.2.2.3ProbabilityofHavingaLandslide
SusceptibleDeposit.........................................435
4.2.2.2.4PermanentGroundDisplacements....................435
4.2.2.3SurfaceFaultRupture..........................................................437
4.2.2.3.1PermanentGroundDisplacements....................437
4.2.3 GuidanceforExpertGeneratedGroundFailureEstimation...........438
4.2.3.1InputRequirementsandOutputInformation.....................438
4.2.3.1.1Liquefaction........................................................438
4.2.3.1.2Landsliding..........................................................439
4.2.3.1.3SurfaceFaultRupture.........................................439
4.2.3.2Liquefaction.........................................................................440
4.2.3.2.1Background.........................................................440
4.2.3.2.2Susceptibility......................................................442
4.2.3.2.3LiquefactionProbability.....................................443
4.2.3.2.4PermanentGroundDisplacement......................444
4.2.3.3Landsliding...........................................................................445
4.2.3.3.1Background.........................................................445
4.2.3.3.2LandslideSusceptibility......................................445
4.2.3.3.3ProbabilityofLandsliding...................................446
4.2.3.3.4PermanentGroundDisplacements....................446
4.2.3.4SurfaceFaultRupture..........................................................446
4.2.3.4.1PermanentGroundDisplacements....................446
4.3 References.....................................................................................................448

5. DirectPhysicalDamageGeneralBuildingStock
5.1 Introduction....................................................................................................51
5.1.1 Scope51
5.1.2 InputRequirementsandOutputInformation...................................53
5.1.3 FormofDamageFunctions................................................................53
5.2 DescriptionofModelBuildingTypes..............................................................57
5.2.1 StructuralSystems.............................................................................58
5.2.2 NonstructuralComponents.............................................................512
5.3 DescriptionofBuildingDamageStates.........................................................513
5.3.1 StructuralDamage...........................................................................515
5.3.2 NonstructuralDamage.....................................................................523
5.4 BuildingDamageDuetoGroundShaking.....................................................524
5.4.1 Overview..........................................................................................524
5.4.2 CapacityCurves................................................................................526
5.4.3 FragilityCurves.................................................................................537
5.4.3.1 Background.........................................................................538
5.4.3.2 DevelopmentofDamageStateMedians............................539
5.4.3.3 DevelopmentofDamageStateVariability..........................540
5.4.3.4 StructuralDamage..............................................................542
5.4.3.5 NonstructuralDamageDriftSensitive
Components........................................................................548
5.4.3.6 NonstructuralDamageAccelerationSensitiveComponents554
5.4.4 StructuralFragilityCurvesEquivalentPeakGroundAcceleration.560
5.5 BuildingDamageDuetoGroundFailure......................................................568
5.5.1 Overview..........................................................................................568
5.5.2 FragilityCurvesPeakGroundDisplacement..................................568
5.6 EvaluationofBuildingDamage.....................................................................569
5.6.1 Overview..........................................................................................569
5.6.2 DamageDuetoGroundShaking......................................................569
5.6.2.1DemandSpectrumReductionforEffective
Damping....................................................................................569
5.6.2.2 DamageStateProbability...................................................573
5.6.3 GroundDamageduetoCombinedGroundFailureand
GroundShaking................................................................................575
5.6.4 CombinedDamagetoOccupancyClasses.......................................576
5.7 GuidanceforExpertUsers............................................................................577
5.7.1 SelectionofRepresentativeSeismicDesignLevel...........................577
5.7.2 DevelopmentofDamageFunctionsforOtherBuildings.................579
5.8 BuildingDamageReferences........................................................................580
6. DirectPhysicalDamageEssentialandHighPotentialLossFacilities
6.1 Introduction....................................................................................................61
6.1.1 Scope61
6.1.2 EssentialFacilitiesClassification........................................................61
6.1.3 InputRequirementsandOutputInformation...................................63
6.1.4 FormofDamageFunctions................................................................64
6.2 DescriptionofModelBuildingTypes..............................................................65
6.3 DescriptionofBuildingDamageStates...........................................................66
6.4 BuildingDamageDuetoGroundShakingSpecialBuildings.........................66
6.4.1 Overview............................................................................................66
6.4.2 CapacityCurvesSpecialBuildings....................................................67
6.4.3 FragilityCurvesSpecialBuildings...................................................612
6.4.3.1Background.........................................................................612
6.4.3.2StructuralDamageSpecialBuildings................................612
6.4.3.3NonstructuralDamageDriftSensitive..............................615
6.4.3.4NonstructuralDamageAcceleration
SensitiveComponents...............................................................620
6.4.4 StructuralFragilityCurvesEquivalentPGA....................................624
6.5 DamageDuetoGroundFailureSpecialBuildings......................................628
6.6 EvaluationofBuildingDamageEssentialFacilities.....................................628
6.6.1 Overview..........................................................................................628
6.6.2 DamageDuetoGroundShaking......................................................628
6.6.2.1DemandSpectrumReductionforEffective
DampingSpecialBuildings.......................................................629
6.6.2.2DamageStateProbability...................................................630
6.6.3 CombinedDamageDuetoGroundFailureandGround
Shaking631
6.6.4 CombinedDamagetoEssentialFacilities........................................632
6.7 GuidanceforExpertUsers............................................................................632
6.7.1 SelectionofRepresentativeSeismicDesignLevel...........................632
6.7.2 HighPotentialLossFacilities............................................................633
6.7.2.1Introduction........................................................................633
6.7.2.2InputRequirementandOutputInformation......................633
6.7.2.3FormofDamageFunctionandDamageEvaluation...........634
6.8 EssentialFacilityandHPLDamageReferences.............................................634
6.9 RestorationCurves........................................................................................634
7. DirectPhysicalDamagetoLifelinesTransportationSystems
7.1 HighwayTransportationSystem.....................................................................71
7.1.1 Introduction.......................................................................................71
7.1.2 Scope71
7.1.3 InputRequirementsandOutputInformation...................................73
7.1.4 FormofDamageFunctions................................................................74
7.1.5 DescriptionofHighwayComponents................................................74
7.1.6 DefinitionofDamageStates..............................................................78
7.1.7 ComponentRestorationCurves.........................................................79
7.1.8 DevelopmentofDamageFunctions.................................................711
7.1.9 GuidanceforLossEstimationUsingAdvancedDataand
Models.............................................................................................717
7.1.10 References.......................................................................................717
FiguresforSection7.1..................................................................................718
7.2 RailwayTransportationSystem....................................................................724
7.2.1 Introduction.....................................................................................724
7.2.2 Scope724
7.2.3 InputRequirementsandOutputInformation.................................725
7.2.4 FormofDamageFunctions..............................................................725
7.2.5 DescriptionofRailwaySystemComponents...................................726
7.2.6 DefinitionofDamageStates............................................................727
7.2.7 ComponentRestorationCurves.......................................................730
7.2.8 DevelopmentofDamageFunctions.................................................732
7.2.9 GuidanceforLossEstimationUsingAdvancedDataand
Models.............................................................................................737
7.2.10 References.......................................................................................737
FiguresforSection7.2..................................................................................738
7.3 LightRailTransportationSystem..................................................................748
7.3.1 Introduction.....................................................................................748
7.3.2 Scope748
7.3.3 InputRequirementsandOutputInformation.................................748
7.3.4 FormofDamageFunctions..............................................................749
7.3.5 DescriptionofLightRailSystemComponents.................................750
7.3.6 DefinitionofDamageStates............................................................750
7.3.7 ComponentRestorationCurves.......................................................752
7.3.8 DevelopmentofDamageFunctions.................................................752
7.3.9 GuidanceforLossEstimationUsingAdvancedDataand
Models.............................................................................................753
7.3.10 References.......................................................................................753
FiguresforSection7.3..................................................................................754
7.4 BusTransportationSystem...........................................................................755
7.4.1 Introduction.....................................................................................755
7.4.2 Scope755
7.4.3 InputRequirementsandOutputInformation.................................755
7.4.4 FormofDamageFunctions..............................................................756
7.4.5 DescriptionofBusSystemComponents..........................................757
7.4.6 DefinitionofDamageStates............................................................757
7.4.7 ComponentRestorationCurves.......................................................758
7.4.8 DevelopmentofDamageFunctions.................................................758
7.4.9 GuidanceforLossEstimationUsingAdvancedDataand
Models.............................................................................................759
7.4.10 References.......................................................................................760
7.5 PortTransportationSystem..........................................................................761
7.5.1 Introduction.....................................................................................761
7.5.2 Scope761
7.5.3 InputRequirementsandOutputInformation.................................761
7.5.4 FormofDamageFunctions..............................................................762
7.5.5 DescriptionofPortComponents.....................................................763
7.5.6 DefinitionofDamageStates............................................................763
7.5.7 ComponentRestorationCurves.......................................................765
7.5.8 DevelopmentofDamageFunctions.................................................766
7.5.9 GuidanceforLossEstimationUsingAdvancedDataand
Models.............................................................................................767
7.5.10 References.......................................................................................768
FiguresforSection7.5..................................................................................769
7.6 FerryTransportationSystem........................................................................773
7.6.1 Introduction.....................................................................................773
7.6.2 Scope773
7.6.3 InputRequirementsandOutputInformation.................................774
7.6.4 FormofDamageFunctions..............................................................774
7.6.5 DescriptionofFerrySystemComponents.......................................775
7.6.6 DefinitionofDamageStates............................................................775
7.6.7 ComponentRestorationCurves.......................................................777
7.6.8 DevelopmentofDamageFunctions.................................................777
7.6.9 GuidanceforLossEstimationUsingAdvancedDataand
Models.............................................................................................777
7.6.10 References.......................................................................................777
7.7 AIRPORTTRANSPORTATIONSYSTEM...........................................................778
7.7.1 Introduction.....................................................................................778
7.7.2 Scope778
7.7.3 InputRequirementsandOutputInformation.................................778
7.7.4 FormofDamageFunctions..............................................................779
7.7.5 DescriptionofAirportComponents.................................................780
7.7.6 DefinitionofDamageStates............................................................780
7.7.7 ComponentRestorationCurves.......................................................782
7.7.8 DevelopmentofDamageFunctions.................................................783
7.7.9 GuidanceforLevel3LossEstimation..............................................784
7.7.10 References.......................................................................................784
FiguresforSection7.7..................................................................................785
Appendix7A ................................................................................................787
Appendix7B ................................................................................................788
Appendix7C ................................................................................................792
Appendix7D ................................................................................................793
8. DirectDamagetoLifelinesUtilitySystems
8.1 POTABLEWATERSYSTEMS.............................................................................81
8.1.1 Introduction.......................................................................................81
8.1.2 Scope81
8.1.3 InputRequirementsandOutputInformation...................................83
8.1.4 FormofDamageFunctions................................................................84
8.1.5 DescriptionofPotableWaterSystemComponents..........................85
8.1.6 DefinitionofDamageStates..............................................................87
8.1.6.1DamageStateDefinitionsforComponentsOther
thanPipelines..............................................................................87
8.1.6.2 DefinitionofDamageStatesforPipelines............................89
8.1.7 ComponentRestorationCurves.........................................................89
8.1.8 DevelopmentofDamageFunctions.................................................812
8.1.9 SystemPerformance........................................................................818
8.1.10 GuidanceforLossEstimationUsingAdvancedData
andModels......................................................................................819
8.1.11 References.......................................................................................820
FiguresforSection8.1..................................................................................821
8.2 WASTEWATERSYSTEM................................................................................834
8.2.1 Introduction.....................................................................................834
8.2.2 Scope834
8.2.3 InputRequirementsandOutputInformation.................................834
8.2.4 FormofDamageFunctions..............................................................834
8.2.5 DescriptionofWasteWaterSystemComponents..........................835
8.2.6 DefinitionofDamageStates............................................................836
8.2.6.1DamagesStatesDefinitionsforComponentsother
thanSewers/Interceptors..........................................................836
8.2.6.2 DamageStatesDefinitionsforSewers/Interceptors..........837
8.2.7 ComponentRestorationCurves.......................................................837
8.2.8 DevelopmentofDamageFunctions.................................................838
8.2.9 GuidanceforLossEstimationUsingAdvancedDataand
Models.............................................................................................840
8.2.10 References.......................................................................................841
FiguresforSection8.2..................................................................................842
8.3 OILSYSTEMS 846
8.3.1 Introduction.....................................................................................846
8.3.2 Scope846
8.3.3 InputRequirementsandOutputInformation.................................846
8.3.4 FormofDamageFunctions..............................................................847
8.3.5 DescriptionofOilSystemComponents...........................................847
8.3.6 DefinitionofDamageStates............................................................848
8.3.6.1DamagestatesdefinitionsforComponentsother
thanPipelines............................................................................848
8.3.6.2 DamageStateDefinitionsforPipelines..............................849
8.3.7 ComponentRestorationCurves.......................................................850
8.3.8 DevelopmentofDamageFunctions.................................................850
8.3.9 GuidanceforLossEstimationUsingAdvancedDataand
Models.............................................................................................853
8.3.10 References.......................................................................................853
FiguresforSection8.3..................................................................................854
8.4 NATURALGASSYSTEMS................................................................................860
8.4.1 Introduction.....................................................................................860
8.4.2 Scope860
8.4.3 InputRequirementsandOutputInformation.................................860
8.4.4 FormofDamageFunctions..............................................................861
8.4.5 DescriptionofNaturalGasSystemComponents.............................861
8.4.6 DefinitionofDamageStates............................................................861
8.4.6.1DamageStatesDefinitionsforCompressor
Stations..............................................................................862
8.4.6.2 DamageStatesDefinitionsforPipelines.............................862
8.4.7 ComponentRestorationCurves.......................................................862
8.4.8 DevelopmentofDamageFunctions.................................................862
8.4.9 GuidanceforLossEstimationUsingAdvancedDataand
Models.............................................................................................863
8.4.10 References.......................................................................................863
8.5 ELECTRICPOWERSYSTEM.............................................................................864
8.5.1 Introduction.....................................................................................864
8.5.2 Scope864
8.5.3 InputRequirementsandOutputInformation.................................864
8.5.4 FormofDamageFunctions..............................................................865
8.5.5 DescriptionofElectricPowerSystemComponents.........................865
8.5.6 DefinitionofDamageStates............................................................866
8.5.7 ComponentRestorationCurves.......................................................868
8.5.8 DevelopmentofDamageFunctions.................................................869
8.5.9 PowerOutage&PerformanceEvaluationforElectric
PowerSystems.................................................................................872
8.5.10 GuidanceforLossEstimationUsingAdvancedDataand
Models.............................................................................................874
8.5.11 References.......................................................................................875
FiguresforSection8.5..................................................................................876
8.6 COMMUNICATIONSYSTEMS.........................................................................884
8.6.1 Introduction.....................................................................................884
8.6.2 Scope884
8.6.3 InputRequirementsandOutputInformation.................................884
8.6.4 FormofDamageFunctions..............................................................885
8.6.5 DescriptionofCommunicationSystemComponents......................885
8.6.6 DefinitionofDamageStates............................................................885
8.6.7 ComponentRestorationCurves.......................................................886
8.6.8 DevelopmentofDamageFunctions.................................................887
8.6.9 GuidanceforLossEstimationUsingAdvancedDataand
Models.............................................................................................887
8.6.10 References.......................................................................................888
FiguresforSection8.6..................................................................................889
Appendix8A ................................................................................................892
Appendix8B ................................................................................................897
Appendix8C ................................................................................................899
Appendix8D ..............................................................................................8102
Appendix8E ..............................................................................................8107
9. InducedDamageModelsInundation
9.1 Introduction....................................................................................................91
9.1.1 Scope91
9.1.2 FormofInundationEstimate.............................................................91
9.1.3 InputRequirementsandOutputInformation...................................93
9.1.3.1InputRequirements...............................................................93
9.1.3.1.1DamFailure..........................................................93
9.1.3.1.2LeveeFailure........................................................94
9.1.3.1.3InputRequirementsTsunami.............................96
9.1.3.1.4InputRequirementsSeiche................................96
9.1.3.2OutputInformation................................................................97
9.2 DescriptionofMethodology...........................................................................97
9.2.1 DamFailure........................................................................................97
9.2.2 LeveeFailure......................................................................................98
9.2.3 Tsunami..............................................................................................98
9.2.4 Seiche.................................................................................................98
9.3 GuidanceforExpertGeneratedEstimates.....................................................98
9.3.1 DamFailure........................................................................................99
9.3.2 LeveeFailure....................................................................................912
9.3.3 Tsunami............................................................................................912
9.3.4 Seiche...............................................................................................913
9.4 InundationReferences..................................................................................914
10. InducedDamageModelsFireFollowingEarthquake
10.1 Introduction..................................................................................................101
10.1.1 Scope101
10.1.2 FormofDamageEstimates..............................................................103
10.1.3 InputRequirements.........................................................................103
10.2 DescriptionofMethodology.........................................................................104
10.2.1 Ignition.............................................................................................104
10.2.2 Spread............................................................................................1010
10.2.3 Suppression....................................................................................1014
10.2.3.1DiscoveryTime.................................................................1014
10.2.3.2ReportTime.....................................................................1014
10.2.3.3ArrivalTime......................................................................1015
10.2.3.4ControlTime.....................................................................1016
10.2.3.4.1RoomandContentFires.................................1016
10.2.3.4.2StructureFireEnginesNeeded.....................1016
10.2.3.4.3StructureFireWaterNeeded........................1017
10.2.3.4.4EnginesAvailable............................................1019
10.2.3.4.5WaterAvailable..............................................1020
10.2.3.4.6FireSpreadwithPartiallyEffective
Suppression..................................................1020
10.2.3.4.7FireSpreadatNaturalFireBreaks..................1021
10.3 GuidanceforExpertGeneratedEstimates.................................................1022
10.4 References...................................................................................................1023
11. InducedDamageModelsHazardousMaterialsRelease
11.1 Introduction..................................................................................................111
11.1.1Scope..................................................................................................111
11.1.2ClassificationofHazardousMaterials................................................113
11.1.3InputRequirementsandOutputInformation....................................113
11.2 DescriptionofMethodology.........................................................................116
11.3 GuidanceforExpertGeneratedEstimates...................................................116
11.4 References.....................................................................................................119
Appendix11A 1112
Appendix11B 1118

12. InducedDamageMethodsDebris
12.1 Introduction..................................................................................................121
12.1.1 Scope121
12.1.2 FormofDamageEstimate................................................................121
12.1.3 InputRequirementsandOutputInformation.................................121
12.2 DescriptionofMethodology.........................................................................123
12.2.1DebrisGeneratedFromDamagedBuildings......................................123
12.3 GuidanceforExpertGeneratedEstimates...................................................128
12.4 References.....................................................................................................128
13. DirectSocialLossesCasualties
13.1 Introduction..................................................................................................131
13.1.1 Scope131
13.1.2 FormofCasualtyEstimate...............................................................133
13.1.3 InputRequirements.........................................................................134
13.2 DescriptionofMethodology.......................................................................1319
13.2.1 EarthquakeCasualtyModel...........................................................1319
13.2.2 AlternativeEstimationofCasualtyRates.......................................1321
13.2.3 CasualtiesduetoOutdoorFallingHazards....................................1322
13.2.4 CasualtyRatesResultingfromBridgeCollapse..............................1323
13.3 References...................................................................................................1324

14. DirectSocialLossesDisplacedHouseholdsDuetoLossofHousing
HabitabilityandShortTermShelterNeeds
14.1 Introduction..................................................................................................141
14.1.1 Scope141
14.2 DisplacedHouseholdsFormofLossEstimate............................................143
14.2.1 InputRequirementsDisplacedHouseholds...................................143
14.2.2 DescriptionofMethodology............................................................143
14.3 ShortTermShelterNeedsFormofLossEstimate......................................144
14.3.1 InputRequirementsShortTermShelterNeeds.............................145
14.3.2 DescriptionofMethodology............................................................145
14.3.3 UserdefinedChangestoWeightandModificationFactors............147
14.3.4 GuidanceforEstimatesUsingAdvancedDataandModels.............148
14.4 GuidanceforEstimatingLongTermHousingRecovery...............................148
14.5 References.....................................................................................................149
15. DirectEconomicLosses
15.1 Introduction..................................................................................................151
15.1.1 Scope154
15.1.2 FormofDirectEconomicLossEstimates.........................................155
15.1.3 InputRequirements.........................................................................156
15.2 DescriptionofMethodology:Buildings.........................................................157
15.2.1 BuildingRepairandReplacementCosts..........................................157
15.2.1.1DefaultValuesforBuildingRepairCosts.........................1510
15.2.1.2ProcedureforUpdatingBuildingReplacementCostEstimates
.......................................................................................................1514
15.2.2 BuildingContentsLosses................................................................1514
15.2.3 BusinessInventoryLosses..............................................................1516
15.2.4 BuildingRepairTime/LossofFunction..........................................1518
15.2.5 RelocationExpenses......................................................................1522
15.2.6LossofIncome................................................................................1526
15.2.6.1RecaptureFactors............................................................1526
15.2.7 RentalIncomeLosses.....................................................................1528
15.2.8 GuidanceforEstimateUsingAdvancedDataandModelsAnalysis1528
15.3 DescriptionofMethodology:Lifelines........................................................1530
15.3.1 TransportationSystems.................................................................1536
15.3.1.1HighwaySystems.............................................................1536
15.3.1.2RailwaySystems...............................................................1537
15.3.1.3LightRailSystems.............................................................1538
15.3.1.4BusSystems......................................................................1539
15.3.1.5PortSystems....................................................................1539
15.3.1.6FerrySystems...................................................................1540
15.3.1.7AirportSystems................................................................1541
15.3.2 UtilitySystems................................................................................1542
15.3.2.1PotableWaterSystems..................................................1542
15.3.2.2WasteWaterSystems....................................................1543
15.3.2.3OilSystems.....................................................................1544
15.3.2.4NaturalGasSystems.......................................................1545
15.3.2.5ElectricPowerSystems...................................................1545
15.3.2.6CommunicationSystems................................................1545
15.4 References...................................................................................................1546
Appendix15A ..............................................................................................1547
Appendix15B ..............................................................................................1551
16. IndirectEconomicLosses
16.1 Introduction..................................................................................................161
16.2 WhatareIndirectLosses?.............................................................................161
16.2.1 SupplyShortagesandForwardLinkedLosses.................................163
16.2.2 DemandEffectsandBackwardLinkedLosses.................................164
16.2.3 Regionalvs.NationalLosses............................................................164
16.3 InterindustryModels....................................................................................165
16.3.1 APrimeronInputOutputTechniques.............................................166
16.3.2 AnIllustrationofBackwardLinkedLosses.......................................169
16.3.3 TheImpactofOutsideReconstructionAidonthe
RegionandtheNation...................................................................1610
16.4 TheIndirectLossModule............................................................................1611
16.4.1 DamageLinkagetotheDirectLossModule..................................1612
16.4.2 SupplySideAdjustmentsandRebalancingtheEconomy..............1614
16.4.3 TheTimeDimension......................................................................1618
16.4.4 TheEffectofRebuildingandBorrowing........................................1618
16.4.5TheIssueofAggregation................................................................1619
16.5 RunningtheModel......................................................................................1619
16.5.1DefaultDataAnalysisInputs,OperationandOutput.....................1620
16.5.1.1UserInputandDefaultData...........................................1620
16.5.1.2CalculationofIndirectLoss.............................................1623
16.5.1.3TheFormatoftheOutput...............................................1625
16.5.2 UserSuppliedDataAnalysis..........................................................1626
16.5.2.1IMPLANInputOutputData............................................1626
16.5.2.2SpecifyingIndirectLossFactors......................................1628
16.5.3 AdvancedDataandModelsAnalysis...........................................1632
16.5.3.1ExtendingtheIndirectLossModule...............................1632
16.5.3.2AlternativeModelingTechniques...................................1634
16.6 ExampleSolutions.......................................................................................1635
16.6.1SimpleOneSectorSupplyShockNoExcessCapacity...................1635
16.6.2TheNorthridgeEarthquake............................................................1635
16.6.3TheSensitivityofIndirectLosstoCapacity,Damage
andReconstruction........................................................................1642
16.6.4ObservationaboutIndirectLoss......................................................1654
16.7 References...................................................................................................1655
Appendix16A ..............................................................................................1659
17.AnnualizedLosses
16.1 Introduction..................................................................................................171
16.2 AverageAnnualizedEarthquakeLossComputation.....................................172

MESSAGETOUSERS

TheHazusEarthquakeModelisdesignedtoproducelossestimatesforusebyfederal,state,
regionalandlocalgovernmentsinplanningforearthquakeriskmitigation,emergency
preparedness,responseandrecovery.Themethodologydealswithnearlyallaspectsofthebuilt
environment,andawiderangeofdifferenttypesoflosses.Extensivenationaldatabasesare
embeddedwithinHazus,containinginformationsuchasdemographicaspectsofthepopulation
inastudyregion,squarefootagefordifferentoccupanciesofbuildings,andnumbersand
locationsofbridges.Embeddedparametershavebeenincludedasneeded.Usingthis
information,userscancarryoutgenerallossestimatesforaregion.TheHazusmethodologyand
softwareareflexibleenoughsothatlocallydevelopedinventoriesandotherdatathatmore
accuratelyreflectthelocalenvironmentcanbesubstituted,resultinginincreasedaccuracy.
Uncertaintiesareinherentinanylossestimationmethodology.Theyariseinpartfrom
incompletescientificknowledgeconcerningearthquakesandtheireffectsuponbuildingsand
facilities.Theyalsoresultfromtheapproximationsandsimplificationsthatarenecessaryfor
comprehensiveanalyses.Incompleteorinaccurateinventoriesofthebuiltenvironment,
demographicsandeconomicparametersaddtotheuncertainty.Thesefactorscanresultina
rangeofuncertaintyinlossestimatesproducedbytheHazusEarthquakeModel,possiblyatbest
afactoroftwoormore.
Themethodologyhasbeentestedagainstthejudgmentofexpertsand,totheextentpossible,
againstrecordsfromseveralpastearthquakes.However,limitedandincompletedataabout
actualearthquakedamageprecludescompletecalibrationofthemethodology.Nevertheless,
whenusedwithembeddedinventoriesandparameters,theHazusEarthquakeModelhas
providedacredibleestimateofsuchaggregatedlossesasthetotalcostofdamageandnumbers
ofcasualties.TheEarthquakeModelhasdonelesswellinestimatingmoredetailedresults
suchasthenumberofbuildingsorbridgesexperiencingdifferentdegreesofdamage.Such
resultsdependheavilyuponaccurateinventories.TheEarthquakeModelassumesthesamesoil
conditionforalllocations,andthishasprovedsatisfactoryforestimatingregionallosses.Of
course,thegeographicdistributionofdamagemaybeinfluencedmarkedlybylocalsoil
conditions.InthefewinstanceswheretheEarthquakeModelhasbeenpartiallytestedusing
actualinventoriesofstructurespluscorrectsoilsmaps,ithasperformedreasonablywell.
Usersshouldbeawareofthefollowingspecificlimitations:
WhiletheHazusEarthquakeModelcanbeusedtoestimatelossesforanindividual
building,theresultsmustbeconsideredasaverageforagroupofsimilarbuildings.Itis
frequentlynotedthatnominallysimilarbuildingshaveexperiencedvastlydifferent
damageandlossesduringanearthquake.

Whenusingembeddedinventories,accuracyoflossesassociatedwithlifelinesmaybe
lessthanforlossesfromthegeneralbuildingstock.Theembeddeddatabasesand
assumptionsusedtocharacterizethelifelinesystemsinastudyregionarenecessarily
incompleteandoversimplified.
Basedonseveralinitialstudies,thelossesfromsmallmagnitudeearthquakes(lessthan
M6.0)centeredwithinanextensiveurbanregionappeartobeoverestimated.
BecauseofapproximationsinmodelingoffaultsinCalifornia,theremaybe
discrepanciesinmotionspredictedwithinsmallareasimmediatelyadjacenttofaults.
Thereisconsiderableuncertaintyrelatedtothecharacteristicsofgroundmotioninthe
EasternU.S.TheembeddedattenuationrelationsintheEarthquakeModel,whichare
thosecommonlyrecommendedfordesign,tendtobeconservative.Henceuseofthese
relationsmayleadtooverestimationoflossesinthisregion,bothforscenarioevents
andwhenusingprobabilisticgroundmotion.
Hazusshouldstillberegardedasaworkinprogress.Additionaldamageandlossdatafrom
actualearthquakesandfurtherexperienceinusingthesoftwarewillcontributeto
improvementsinfuturereleases.ToassistusinfurtherimprovingHazus,usersareinvitedto
submitcommentsonmethodologicalandsoftwareissuesbyletter,faxoremailto:

David Adler
Zimmerman Associates, Inc
7390 Coca Cola Drive.
Hanover, MD 21076
Tel: 410-712-7401
Fax: 800-358-9620
E-Mail: david.adler@riskmapcds.com


Eric Berman
Department of Homeland Security
Federal Emergency Management Agency
Mitigation Division
500 C Street, S.W.
Washington, DC 20472
Tel: 202-646-3427
Fax: 202-646-2787
E-Mail: Eric.Berman@dhs.gov

WHATSNEWINHazus

EARTHQUAKEMODEL

PleaserefertotheHazusMHEarthquakeUsersManualforahistoryofthefeaturesforthe
differentversionsofHazusMH.

TheUsersManualhasalsodetailsabouttheinstallationofthesoftware,itslimitationsand
capabilities,andinformationonhowtoobtaintechnicalsupport.
11
HazusMHTechnicalManual
Chapter 1
Introduction to the FEMA Loss Estimation Methodology


1.1 Background

The Technical Manual describes the methods for performing earthquake loss estimation.
It is based on a multi-year project to develop a nationally applicable methodology for
estimating potential earthquake losses on a regional basis. The project has being
conducted for the National Institute of Building Science (NIBS) under a cooperative
agreement with the Federal Emergency Management Agency (FEMA).

The primary purpose of the project is to develop guidelines and procedures for making
earthquake loss estimates at a regional scale. These loss estimates would be used
primarily by local, state and regional officials to plan and stimulate efforts to reduce risks
from earthquakes and to prepare for emergency response and recovery. A secondary
purpose of the project is to provide a basis for assessing nationwide risk of earthquake
losses.

The methodology development and software implementation has been performed by a
team of earthquake loss experts composed of earth scientists, engineers, architects,
economists, emergency planners, social scientists and software developers. The
Earthquake Committee has provided technical direction and review of work with
guidance from the Project Oversight Committee (POC), a group representing user
interests in the earthquake engineering community.

1.2 Technical Manual Scope

The scope of the Technical Manual includes documentation of all methods and data that
are used by the methodology. Loss estimation methods and data are obtained from
referenced sources tailored to fit the framework of the methodology, or from new
methods and data developed when existing methods and data were lacking or were not
current with the state of the art.

The Technical Manual is a comprehensive, highly technical collection of methods and
data covering a broad range of topics and disciplines, including earth science,
seismic/structural engineering, social science and economics. The Technical Manual is
written for readers who are expected to have some degree of expertise in the technical
topic of interest, and may be inappropriate for readers who do not have this background.

As described in Chapter 2, a separate User Manual describes the earthquake loss
estimation methodology in non-technical terms and provides guidance to users in the
application of the methodology. The methodology software is implemented using
Geographical Information System (GIS) software as described in the Technical Manual.

1.3 Technical Manual Organization
12
Chapter1IntroductiontotheFEMALossEstimationMethodology

The Technical Manual contains sixteen chapters. Chapter 2 describes the overall
framework of the methodology and provides background on the approach developed used
to meet the projects objectives. Chapter 3 discusses inventory data, including
classification schemes of different systems, attributes required to perform damage and
loss estimation, and the data supplied with the methodology. Sources and methods of
collection of inventory data are not covered in Chapter 3, but may be found in the User
Manual.

Chapters 4 through 16 cover, respectively, each of thirteen major components or
subcomponents (modules) of the methodology. Each of the major components and
subcomponents are described in Chapter 2. A flowchart is provided in Chapter 2 as a
"road map" of the relationships between modules of the methodology. This flowchart is
repeated at the beginning of each chapter with the module of interest high-lighted to show
input from and output to other modules of the methodology.
21

HazusMHTechnicalManual
Chapter 2
Overall Approach and Framework of Methodology

This chapter describes the overall approach used by the developers to meet the objectives
of the project, the components and subcomponents of earthquake loss estimation and
their relationship within the framework of methodology.

2.1 Vision Statement

The overall approach for the project is based on the following "vision" of the earthquake
loss estimation methodology.

The earthquake loss estimation methodology will provide local, state and regional
officials with the tools necessary to plan and stimulate efforts to reduce risk from
earthquakes and to prepare for emergency response and recovery from an
earthquake. The methodology will also provide the basis for assessment of
nationwide risks of earthquake loss.

The methodology can be used by a variety of users with needs ranging from
simplified estimates that require minimal input to refined calculations of
earthquake loss. The methodology may be implemented using either
geographical information system (GIS) technology provided in a software
package or by application of the theory documented in a Technical Manual. An
easily understood User Manual will guide implementation of the methodology by
either technical or non-technical users.

The vision of earthquake loss estimation requires a methodology that is both flexible,
accommodating the needs of a variety of different users and applications, and able to
provide the uniformity of a standardized approach. The framework of the methodology
includes each of the components shown in Figure 2-1: Potential Earth Science Hazard
(PESH), Inventory, Direct Physical Damage, Induced Physical Damage, Direct
Economic/Social Loss and Indirect Economic Loss. As indicated by arrows in the figure,
modules are interdependent with output of some modules acting as input to others. In
general, each of the components will be required for loss estimation. However, the
degree of sophistication and associated cost will vary greatly by user and application. It
is therefore necessary and appropriate that components have multiple levels (multiple
modules) of detail or precision when required to accommodate user needs.

Framing the earthquake loss estimation methodology as a collection of modules permits
adding new modules (or improving models/data of existing modules) without reworking
the entire methodology. Improvements may be made to adapt modules to local or
regional needs or to incorporate new models and data. The modular nature of the
methodology permits a logical evolution of the methodology as research progresses and
the state-of-the-art advances.


22
Chapter2OverallApproachandFrameworkofMethodology

8. Lifelines-
Utility
Systems
4. Ground Motion 4. Ground Failure
Direct Physical
Damage
6. Essential and
High Potential
Loss Facilities
12. Debris 10. Fire 15. Economic 14. Shelter 9. Inundation 11. HazMat
16. Indirect
Economic
Losses
Potential Earth Science Hazards
Direct Economic/
Social Losses
Induced Physical
Damage
7. Lifelines-
Transportation
Systems
5. General
Building
Stock
13. Casualities


Figure 2.1 Flowchart of the Earthquake Loss Estimation Methodology.

23

HazusMHTechnicalManual
Most users will implement the methodology using the GIS-based software application
provided by NIBS. After initial inventory entry, the program will run efficiently on
desktop computer. The GIS technology provides a powerful tool for displaying outputs
and permits users to "see" the effects of different earthquake scenarios and assumptions.
A User Manual will guide users in program manipulation, input of new data, and changes
to existing data.

Certain users may not wish to use the software application, or may want to augment the
results with supplementary calculations. In such cases, users can refer to the Technical
Manual for a complete description of models and data of each module. The Technical
Manual is useful to technical experts, such as those engineers and scientists that have
conducted previous earthquake loss studies, but might be inappropriate for non-technical
users.

Both technical and non-technical users are guided in the application of the methodology
by the User Manual, which addresses important implementation issues, such as:

(1) Selection of scenario earthquakes and PESH inputs
(2) Selection of appropriate methods (modules) to meet different user needs
(3) Collection of required inventory data, i.e., how to obtain necessary information
(4) Costs associated with inventory collection and methodology implementation
(5) Presentation of results including appropriate terminology, etc.
(6) Interpretation of results including consideration of model/data uncertainty.

The three project deliverables are shown in Figure 2.2.


Figure 2.2 Project Deliverables.
Technical Manual
HAZUS
Software Application
V
olum
e I
V
olum
e II
V
olum
e III
H
A
Z
U
S
H
A
Z
U
S
H
A
Z
U
S
Technical Manual
HAZUS
Software Application
HAZUS
Software Application
V
olum
e I
V
olum
e II
V
olum
e III
H
A
Z
U
S
H
A
Z
U
S
H
A
Z
U
S
Users Manual
H
A
Z
U
S
V
o
lu
m
e
I
Users Manual
H
A
Z
U
S
V
o
lu
m
e
I
Technical Manual
HAZUS
Software Application
HAZUS
Software Application
V
olum
e I
V
olum
e II
V
olum
e III
H
A
Z
U
S
H
A
Z
U
S
H
A
Z
U
S
Technical Manual
HAZUS
Software Application
HAZUS
Software Application
V
olum
e I
V
olum
e II
V
olum
e III
H
A
Z
U
S
H
A
Z
U
S
H
A
Z
U
S
Users Manual
H
A
Z
U
S
V
o
lu
m
e
I
Users Manual
H
A
Z
U
S
V
o
lu
m
e
I
Users Manual
H
A
Z
U
S
V
o
lu
m
e
I
Users Manual
H
A
Z
U
S
V
o
lu
m
e
I
H
A
Z
U
S
V
o
lu
m
e
I
24
Chapter2OverallApproachandFrameworkofMethodology
2.2 Project Objectives

The development of an earthquake loss estimation methodology has been defined by the
eight General Objectives outlined in the NIBS/FEMA "Task Plan for Tasks 2 and 5,"
October 18, 1993. The following sections summarize the approach taken to meet each
objective.

Accommodation of User Needs

The methodology utilizes a modular approach with different modules addressing different
user needs. This approach avoids the need to decide on who is the designated user. The
needs of most, if not all, users are accommodated by the flexibility of a modular
approach.

The GIS technology permits easy implementation by users on desktop computers. The
visual display and interactive nature of a GIS application provides an immediate basis for
exchange of information and dialog with end-users of the results. The User Manual
provides appropriate terminology and definitions, and user-oriented descriptions of the
loss estimation process.

State-of-the-Art

The methodology incorporates available state-of-the-art models in the earthquake loss
estimation methodology. For example, ground shaking hazard and related damage
functions are described in terms of spectral response rather than MMI. Modules include
damage loss estimators not previously found in most studies, such as induced damage due
to fire following earthquake and indirect economic loss. A nationally applicable scheme
is developed for classifying buildings, structures and facilities.

Balance

The methodology permits users to select methods (modules) that produce varying degrees
of precision. The User Manual provides guidance to users regarding the selection of
modules that are appropriate for their needs and which have a proper balance between
different components of earthquake loss estimation.

Flexibility in Earthquake Demand

The methodology incorporates both deterministic (scenario earthquake) and probabilistic
descriptions of spectral response. Alternatively, the proposed methodology accepts user-
supplied maps of earthquake demand. The software application is structured to also
accept externally supplied maps of earthquake ground shaking.

"Uncertainty" in earthquake demand due to spatial variability of ground motion is
addressed implicitly by the variability of damage probability matrices (DPM's) or
fragility curves. Uncertainty in earthquake demand due to temporal variability (i.e.,
25

HazusMHTechnicalManual
earthquake recurrence rate) or uncertainty in the magnitude of earthquake selected for
scenario event may be readily evaluated by the users.

Once the data is input into the software application, any number of scenario events can be
evaluated. The User Manual provides guidance for the consideration of uncertainty,
including that associated with earthquake demand.

Uses of Methodology Data

The User Manual provides recommendations for collecting inventory data that will
permit use of the data for non-earthquake purposes. Inventory information will come
from databases supplied with the methodology and/or collected in databases compatible
with the software. Such data will be available to users for other applications.

Accommodation of Different Levels of Funding

The methodology includes modules that permit different levels of inventory collection
and associated levels of funding. For example, the methodology permits simplified
(Default Data Analysis) estimates of damage and loss, using primarily default data
supplied with the software application. These estimates of damage/loss do not require
extensive inventory collection and can be performed on a modest budget. More precise
damage/loss (User-Supplied Data Analysis) estimates require more extensive inventory
information at additional cost to the user. The User Manual provides guidance to users
regarding trade-offs in cost and accuracy of results.

Standardization

The methodology includes standard methods for:

(1) Inventory data collection based on census tract areas
(2) Using database maps of soil type, ground motion, ground failure, etc.
(3) Classifying occupancy of buildings and facilities
(4) Classifying building structure type
(5) Describing damage states
(6) Developing building damage functions
(7) Grouping, ranking and analyzing lifelines
(7) Using technical terminology
(8) Providing output.

Non-Proprietary

The methodology includes only non-proprietary loss estimation methods. The software
application is non-proprietary to the extent permitted by the GIS-software suppliers.

2.3 Description of Loss Estimation Methodology

26
Chapter2OverallApproachandFrameworkofMethodology
The earthquake loss estimation methodology is an improvement over existing regional
loss estimation methodologies, since it more completely addresses regional impacts of
earthquakes that have been omitted or at best discussed in a qualitative manner in
previous studies. Examples of these impacts are service outages for lifelines, estimates of
fire ignitions and fire spread, potential for a serious hazardous materials release incident,
and indirect economic effects. In addition, strength of this methodology is the ability to
readily display inputs and outputs on GIS-based maps that can be overlaid. By
overlaying maps the user is able to experiment with different scenarios and ask "what if"
questions.

As discussed in Section 2.1, the methodology is modular, with different modules
interacting in the calculation of different losses. Figure 2.1 shows each of the modules
and the flow of information among them. It can be seen that, because of the complexity
of earthquake damage and loss estimation, the model is complex. One advantage of the
modularity of the methodology is that it enables users to limit their studies to selected
losses. For example, a user may wish to ignore induced physical damage when
computing direct economic and social losses. This would eliminate the lower left portion
of the flow diagram along with corresponding input requirements. A limited study may
be desirable for a variety of reasons, including budget and inventory constraints, or the
need to obtain answers to very specific questions.

The methodology has been developed with as much capability as possible. However,
there are certain areas where methods are limited. For example, the methodology
calculates potential exposure to flood (e.g., dam break) or fire (following earthquake) in
terms of the fraction of a geographical area that may be flooded or burned, but does not
have methods for rigorous calculation of damage or loss due to flooding or fire.
Consequently, these two potential contributors to the total loss would not be included in
estimates of economic loss, casualties or loss of shelter.

A limiting factor in performing a study and quality of the inventory is the associated cost.
Collection of inventory is without question the most costly part of performing the study.
Furthermore, many municipalities have limited budgets for performing an earthquake
loss estimation study. Thus, the methodology is structured to accommodate different
users with different levels of resources.

While most users will develop a local inventory that best reflects the characteristics of
their region, such as building types and demographics, the methodology is capable of
producing crude estimates of losses based on a minimum of local input. Of course, the
quality and uncertainty of the results is related to the detail of the inventory and the
economic and demographic data provided. Crude estimates would most likely be used
only as initial estimates to determine where more detailed analyses would be warranted.

At the other end of the spectrum, a user may wish to make detailed assessments of
damage to and service outages for lifelines. Detailed analyses of lifelines require
cooperation and input from utilities and transportation agencies. Lifeline systems require
an understanding of the interactions between components and the potential for alternative
27

HazusMHTechnicalManual
pathways when certain components fail. Thus, without cooperation of utilities, the user is
limited in the quality of analysis that can be performed.

The proposed loss estimation methods are capable of providing estimates of damage to
and service outages for lifelines with a minimum of cooperation from lifeline operators.
These estimates, of course, will have a great deal of uncertainty associated with them.
However, they will be useful for planning purposes and for an initial estimate to
determine where more detailed analyses would be warranted. Many lifeline operators
perform their own detailed earthquake loss studies that incorporate detailed models of
their systems.

Three types of analysis are defined to describe implementation of the methodology by
users with different needs and resources. These types and their definitions are somewhat
arbitrary, and the boundaries between the three types are not well defined. The three
types are defined as follows:

Default Data Analysis: This is the simplest type of analysis requiring minimum effort
by the user as it is based mostly on input provided with the methodology (e.g.
census information, broad regional patterns of seismic code adoption and
earthquake resistance of classes of construction, etc.). The user is not expected
to have extensive technical knowledge. While the methods require some user-
supplied input to run, the type of input required could be gathered by contacting
government agencies or by referring to published information. At this level,
estimates will be crude, and will likely be appropriate only as initial loss
estimates to determine where more detailed analyses are warranted.

Some components of the methodology cannot be performed in a Default Data
Analysis since they require more detailed inventory than that provided with the
methodology. The following are not included in the Default Data Analysis:
damage/loss due to liquefaction, landslide or surface fault rupture; damage/loss
due to tsunamis, seiche or dam failure. At this level, the user has the option (not
required) to enter information about hazardous substances and emergency
facilities. One week to a month would be required to collect relevant
information depending on the size of the region and the level of detail the user
desires.

User-Supplied Data Analysis: This type of analysis will be the most commonly used. It
requires more extensive inventory data and effort by the user than Default Data
Analysis. The purpose of this type is to provide the user with the best estimates
of earthquake damage/loss that can be obtained using the standardized methods
of analysis included in the methodology. It is likely that the user will need to
employ consultants to assist in the implementation of certain methods. For
example, a local geotechnical engineer would likely be required to define soil
and ground conditions.

28
Chapter2OverallApproachandFrameworkofMethodology
All components of the methodology can be performed at this level and loss
estimates are based on locally (user) developed inventories. At this level, there
are standardized methods of analysis included in the software, but there is no
standardized User-Supplied Data Analysis study. As the user provides more
complete data, the quality of the analysis and results improve. Depending on
the size of the region and the level of detail desired by the user, one to six
months would be required to obtain the required input for this type of analysis.

Advanced Data and Models Analysis: This type incorporates results from engineering
and economic studies carried out using methods and software not included
within the methodology. At this level, one or more technical experts would be
needed to acquire data, perform detailed analyses, assess damage/loss, and assist
the user in gathering more extensive inventory. It is anticipated that at this level
there will be extensive participation by local utilities and owners of special
facilities. There is no standardized Advanced Data and Models Analysis study.
The quality and detail of the results depend upon the level of effort. Six months
to two years would be required to complete an Advanced Data and Models
Analysis.

To summarize, User-Supplied Data Analysis and Advanced Data and Models Analysis
represent a broad range of analyses, and the line between one type of analysis and another
is fuzzy. The above definitions are provided to understand the scope and flexibility of the
methodology, not to limit its application. The primary limit on the type of analysis will
be the user's ability to provide required data.

Even with perfect data, which can never be obtained, the methodology would not be able
to precisely estimate earthquake loss. Simply put, predictive methods are approximate
and will often have large amounts of uncertainty associated with damage and loss
estimates. A discussion of uncertainty and guidance for users performing earthquake loss
estimation is provided in the User Manual.

31
HazusMHTechnicalManual
Chapter 3
Inventory Data: Collection and Classification


3.1. Introduction

This chapter describes the classification of different buildings and lifeline systems,
data and attributes required for performing damage and loss estimation, and the data
supplied with the methodology. The different systems covered in this chapter include
buildings and facilities, transportation systems, utility systems, and hazardous
material facilities. In addition, census data, county business patterns, and indirect
economic data are discussed. Sources and methods of collecting inventory data can
be found in the Users Manual.

Required input data include both default data (data supplied with the methodology)
and data that must be supplied by the user. Data supplied with the methodology
include default values of classification systems (i.e., mapping relationships) and
default databases (e.g., facility location, census information, and economic factors).
Default data are supplied to assist the user that may not have the resources to develop
inventory data and may be superseded by better information when the user can obtain
such for the study region of interest.

3.2. Direct Damage Data - Buildings and Facilities

This section deals with the general building stock, essential facilities, and high
potential loss facilities.
3.2.1. General Building Stock

The general building stock (GBS) includes residential, commercial, industrial,
agricultural, religious, government, and educational buildings. The damage state
probability of the general building stock is computed at the centroid of the census
tract. The entire composition of the general building stock within a given census tract
is lumped at the centroid of the census tract. The inventory information required for
the analysis to evaluate the probability of damage to occupancy classes is the
relationship between the specific occupancy class and the model building types. This
can be computed directly from the specific occupancy class square footage inventory.

All three models (Earthquake, Wind and Flood) use key common data to ensure that
the users do not have inventory discrepancies when switching from hazard to hazard.
Generally the Earthquake Model and Hurricane display GBS data at the census tract
32
Chapter3InventoryData:CollectionandClassification
level while the Flood Model displays GBS data at the census block.
1
The key GBS
databases include the following:

Square footage by occupancy. These data are the estimated floor area by
specific occupancy (e.g., COM1). For viewing by the user, these data are also
rolled up to the general occupancies (e.g., Residential).

Full Replacement Value by occupancy. These data provide the user with
estimated replacement values by specific occupancy (e.g., RES1). For viewing by
the user, these data are also rolled up to the general occupancies (e.g.,
Commercial).

Building Count by occupancy. These data provide the user with an estimated
building count by specific occupancy (e.g., IND1). For viewing by the user, these
data are also rolled up to the general occupancies (e.g., Government).

General Occupancy Mapping. These data provide a general mapping for the
GBS inventory data from the specific occupancy to general building type (e.g.,
Wood).
2


Demographics. This table provides housing and population statistics for the
study region.

3.2.1.1. Classification

The purpose of a building inventory classification system is to group buildings with
similar damage/loss characteristics into a set of pre-defined building classes. Damage
and loss prediction models can then be developed for model building types which
represent the average characteristics of the total population of buildings within each
class.

The building inventory classification system used in this methodology has been
developed to provide an ability to differentiate between buildings with substantially
different damage and loss characteristics. The following primary parameters
affecting building damage and loss characteristics were given consideration in
developing the building inventory classification system.


1
In order to allow for future alignment between the Hurricane and Flood Models, the Hurricane Model
will display and perform analysis at the census block level if the user has included the Flood Model in
the study region.
2
Generally, all three models will agree, however, a user can modify the general occupancy mapping at
the census block level in the Flood Model thereby requiring them to select an average value at the
tract level in the other two models, which will result in variances. This should not be an issue for users
making this type of change.
33
HazusMHTechnicalManual
Structural parameters affecting structural capacity and response
Basic structural system (steel moment frame)
Building height (low-rise, mid-rise, high-rise)
Seismic design criteria (seismic zone) (Refer to Chapter 5)
Nonstructural elements affecting nonstructural damage
Occupancy (affecting casualties, business interruption and contents damage)
Regional building practices (Refer to Chapter 5)
Variability of building characteristics within the classification

To account for these parameters, the building inventory classification system consists
of a two-dimensional matrix relating building structure (model building) types
grouped in terms of basic structural systems and occupancy classes.

The basic model building types are based on FEMA-178 (FEMA, 1992) building
classes. Building height subclasses were added to reflect the variation of typical
building periods and other design parameters with building height. Mobile homes,
which are not included in the FEMA-178 classification, were also added. A listing of
structural building types, with corresponding labels, descriptions, and heights, is
provided in Table 3.1.

The general building stock is also classified based on occupancy. The occupancy
classification is broken into general occupancy and specific occupancy classes. For
the methodology, the general occupancy classification system consists of seven
groups (residential, commercial, industrial, religion/nonprofit, government, education
and lifelines). There are 33 specific occupancy classes. The building occupancy
classes are given in Table 3.2, where the general occupancy classes are identified in
boldface. The distribution of specific occupancies classes within each general
occupancy class can be computed for each census tract based on the occupancy
square footage inventory (Section 3.6). These relationships are in a form shown in
Table 3A.1 of Appendix 3A.

34
Chapter3InventoryData:CollectionandClassification
Table 3.1: Building Structure (Model Building) Types

Height
No. Label Description Range Typical
Name Stories Stories Feet
1
2
W1
W2
Wood, Light Frame ( 5,000 sq. ft.)
Wood, Commercial and Industrial
(> 5,000 sq. ft.)
1 - 2
All
1
2
14
24
3
4
5
S1L
S1M
S1H
Steel Moment Frame
Low-Rise
Mid-Rise
High-Rise
1 - 3
4 - 7
8+
2
5
13
24
60
156
6
7
8
S2L
S2M
S2H
Steel Braced Frame
Low-Rise
Mid-Rise
High-Rise
1 - 3
4 - 7
8+
2
5
13
24
60
156
9 S3 Steel Light Frame All 1 15
10
11
12
S4L
S4M
S4H
Steel Frame with Cast-in-Place
Concrete Shear Walls
Low-Rise
Mid-Rise
High-Rise
1 - 3
4 - 7
8+
2
5
13
24
60
156
13
14
15
S5L
S5M
S5H
Steel Frame with Unreinforced
Masonry Infill Walls
Low-Rise
Mid-Rise
High-Rise
1 - 3
4 - 7
8+
2
5
13
24
60
156
16
17
18
C1L
C1M
C1H
Concrete Moment Frame
Low-Rise
Mid-Rise
High-Rise
1 - 3
4 - 7
8+
2
5
12
20
50
120
19
20
21
C2L
C2M
C2H
Concrete Shear Walls
Low-Rise
Mid-Rise
High-Rise
1 - 3
4 - 7
8+
2
5
12
20
50
120
22
23
24
C3L
C3M
C3H
Concrete Frame with Unreinforced
Masonry Infill Walls
Low-Rise
Mid-Rise
High-Rise
1 - 3
4 - 7
8+
2
5
12
20
50
120
25 PC1 Precast Concrete Tilt-Up Walls All 1 15
26
27
28
PC2L
PC2M
PC2H
Precast Concrete Frames with
Concrete Shear Walls
Low-Rise
Mid-Rise
High-Rise
1 - 3
4 - 7
8+
2
5
12
20
50
120
29
30
RM1L
RM1M
Reinforced Masonry Bearing
Walls with Wood or Metal Deck
Diaphragms
Low-Rise
Mid-Rise
1-3
4+
2
5
20
50
31
32
33
RM2L
RM2M
RM2H
Reinforced Masonry Bearing
Walls with Precast Concrete
Diaphragms
Low-Rise
Mid-Rise
High-Rise
1 - 3
4 - 7
8+
2
5
12
20
50
120
34
35
URML
URMM
Unreinforced Masonry Bearing
Walls
Low-Rise
Mid-Rise
1 - 2
3+
1
3
15
35
36 MH Mobile Homes All 1 10


35
HazusMHTechnicalManual
Table 3.2: Building Occupancy Classes

Label Occupancy Class Example Descriptions
Residential
RES1 Single Family Dwelling House
RES2 Mobile Home Mobile Home
RES3 Multi Family Dwelling
RES3A Duplex
RES3B 3-4 Units
RES3C 5-9 Units
RES3D 10-19 Units
RES3E 20-49 Units
RES3F 50+ Units
Apartment/Condominium
RES4 Temporary Lodging Hotel/Motel
RES5 Institutional Dormitory Group Housing (military, college), Jails
RES6 Nursing Home
Commercial
COM1 Retail Trade Store
COM2 Wholesale Trade Warehouse
COM3 Personal and Repair Services Service Station/Shop
COM4 Professional/Technical Services Offices
COM5 Banks
COM6 Hospital
COM7 Medical Office/Clinic
COM8 Entertainment & Recreation Restaurants/Bars
COM9 Theaters Theaters
COM10 Parking Garages
Industrial
IND1 Heavy Factory
IND2 Light Factory
IND3 Food/Drugs/Chemicals Factory
IND4 Metals/Minerals Processing Factory
IND5 High Technology Factory
IND6 Construction Office
Agriculture
AGR1 Agriculture
Religion/Non/Profit
REL1 Church/Non-Profit
Government
GOV1 General Services Office
GOV2 Emergency Response Police/Fire Station/EOC
Education
EDU1 Grade Schools
EDU2 Colleges/Universities Does not include group housing
3.2.1.2. Specific Occupancy-to-Model Building Type Mapping
Default mapping schemes for specific occupancy classes (except for RES1) to model
building types by floor area percentage are provided in Tables 3A.2 through 3A.16 of
Appendix 3A. Table 3A.2 through 3A.10 provide the suggested mappings for the
Western U.S. buildings and are based on information provided in ATC-13 (1985).
36
Chapter3InventoryData:CollectionandClassification
Tables 3A.11 through 3A.16 provide the mapping for buildings in the rest of the
United States and are based on proprietary insurance data, opinions of a limited
number of experts, and inferences drawn from tax assessors records. Table 3C.1 in
Appendix 3C provides regional classification of the states. Table 3A.17 through
3A.21 provide model building distribution for the specific occupancy class RES1
on a state-by-state basis. Tables 3A.2 through 3A.10 provide the mapping based on
the height of buildings and the age of construction. The user must provide, for census
tracts on the west coast, the proportion of buildings in low, mid, and high rise
categories, and the proportion of buildings in the three categories according to age
(pre- 1950, 1950-1970, and post 1970). These proportions are used to compute a
weighted sum of matrices in Table 3A.2 through Table 3A.10 to arrive at the default
specific occupancy class to model building type mapping. For the rest of the United
States, Tables 3A.11 through 3A.16 provides the mapping based on the height of
buildings only and the user must provide the proportion of buildings in low-, mid-,
and high-rise categories to compute the default specific occupancy class to model
building type mapping. The default mapping provided in Tables 3A.2 through 3A.16
should be considered as a guide: Accurate mapping may be developed based on the
particular building type distribution within in the study region.

3.2.1.3. The Default General Building Stock Database

The general building stock inventory was developed from the following information:

Census of Population and Housing, 2000: Summary Tape File 1B Extract on CD-
ROM / prepared by the Bureau of Census.

Census of Population and Housing, 2000: Summary Tape File 3 on CD-ROM /
prepared by the Bureau of Census.

Dun & Bradstreet, Business Population Report aggregated by Standard Industrial
Classification (SIC) and Census Block, May 2002.

Department of Energy, Housing Characteristics 1993. Office of Energy Markets
and End Use, DOE/EIA-0314 (93), June 1995.

Department of Energy, A Look at Residential Energy Consumption in 1997,
DOE/EIA-0632(97), November 1999.

Department of Energy, A Look at Commercial Buildings in 1995:
Characteristics, Energy Consumption, and Energy Expenditures, DOE/EIA-0625(95),
October 1998.

The US Census and the Dun & Bradstreet data were used to develop the general
building stock inventory by Census Block and then rolled up to Census Tract. The
three reports from the Department of Energy (DOE) helped in defining regional
variations in characteristics such as number and size of garages, type of foundation,
37
HazusMHTechnicalManual
and number of stories. The inventorys baseline floor area is based on a distribution
contained in the DOEs Energy Consumption Report. An approach was developed
using the same report for determining the valuation of single-family residential homes
by accounting for income as a factor on the cost of housing.

Initially the methodology created the opportunity for the user to develop conflicting
or discrepant square footage totals for single-family residential structures within a
census block between the inventory database and the valuation database. The
solution was to integrate the regional DOE distributions with the income factors
developed for determining valuation. To do this, default values for typical square
footage per single-family home were developed from Energy Information
Administration (EIA) data on heated floor space. These default data, shown in Table
3.3, are provided by region and income group. The breakdown reflects not only how
typical housing size varies across the U.S., but also how in general, higher income
areas tend to contain larger single-family homes.

Consequentially, the default typical square footage data was derived from a detailed,
unpublished database provided by the EIA. Only information on families in single-
family residences, aggregated across all foundation/basement types, was used. The
raw database included information on the number of households by region, income
category, and housing floor space. Regional data were available by 9 multi-state
census divisions (e.g., New England).

The very nature of the default data, both in occupancy classifications and extent of
coverage (national) requires the use of a baseline database collected in a consistent
manner for the nation. The data source changes depending on the general use of the
inventory being explored. For example, to determine the total floor area (square feet)
of single-family residences by census block, one uses a data source like the Census
data. While sufficient for residential occupancy, the Census data does not address
non-residential occupancy classifications.

The development of the default inventory required two major datasets for the two
main elements of the built environment. To create the default inventory for
residential structures, the US Department of Commerces Census of Housing was
used. For commercial and industrial structures, a commercial supplier, Dun &
Bradstreet (D&B) was contacted. The project team performed the aggregation to the
census data, while D&B performed the aggregation to their own data (due to its
proprietary nature).

The STF1B census extract at the census block level allows for the quick
quantification of the single-family residential environment. When combined with the
STF3A census extract at the census block group level, the STF1B can provide a better
proxy of the multi-family environment than using one extract alone. In both the
single-family and multi-family proxies, the proposed methodology represents an
improvement over using single average values similar to the existing Hazus99 data.

38
Chapter3InventoryData:CollectionandClassification
The STF3A extract also provides information that is useful in developing
distributions for the age of buildings within each census block group as well as
valuable demographic data.

The D&B provides a realistic representation of the non-residential environment.
Based on the site specific data contained within their database, D&Bs data is used to
provide a reasonable assessment of the non-residential environment. The processing
of the D&B data is discussed in more detail in Section 3.2.1.4.
3.2.1.4. Specific Occupancy Square Footage

Single-Family Residences (RES1)

The following discussion highlights the data development effort for the RES1 square
foot values by block. The Census Extract STF1B provides estimates of the single
family attached and detached housing units on a block-by-block basis. Several other
sources of information were used to develop distributions of square footage relative to
the income of the census block group. The DOE distributions of income factors was
used to develop a ratio of the census block group income (STF3A field P08A001) and
the average income for the region (the nine multi-state census divisions).

The EIA data provided information regarding the heated floor area in relationship to
income. Income was reported in 25 categories (e.g., $20,000-$22,499) that were
converted into five relative income groups for consistency with the inventory
valuation methodology. Housing floor space data were provided in 7 categories (e.g.,
2,000-2,399 sq. ft.), which, for purposes of computing typical floor space, were
represented by the midpoint of the range (e.g., 2,200 sq. ft.). This enabled average
floor space to be calculated for the 9 census divisions and 5 relative income
categories.

39
HazusMHTechnicalManual
Table 3.3 Typical Square Footage Per Unit (Main Living Area) by
Census Division (R)
1

R = New England
Income Ratio:
Basement
No (j=1) Yes
2
(j=2)
I
k
< 0.5 1300 975
0.5 I
k
< 0.85 1500 1125
0.85 I
k
< 1.25 1800 1350
1.25 I
k
< 2.0 1900 1425
I
k
2.0 2200 1650

R = Middle Atlantic
Income Ratio:
Basement
No (j=1) Yes
2
(j=2)
I
k
< 0.5 1300 975
0.5 I
k
< 0.85 1500 1125
0.85 I
k
< 1.25 1700 1275
1.25 I
k
< 2.0 1900 1425
I
k
2.0 2200 1650

R = East North Central
Income Ratio:
Basement
No (j=1) Yes
2
(j=2)
I
k
< 0.5 1300 975
0.5 I
k
< 0.85 1600 1200
0.85 I
k
< 1.25 1700 1275
1.25 I
k
< 2.0 1800 1350
I
k
2.0 2500 1875

310
Chapter3InventoryData:CollectionandClassification
Table 3.3 Typical Square Footage Per Unit (Main Living Area) by
Census Division (R)
1
(Continued)
R = West North Central
Income Ratio:
Basement
No (j=1) Yes
2
(j=2)
I
k
< 0.5 1300 975
0.5 I
k
< 0.85 1500 1125
0.85 I
k
< 1.25 1800 1350
1.25 I
k
< 2.0 1800 1350
I
k
2.0 2300 1725

R = South Atlantic
Income Ratio:
Basement
No (j=1) Yes
2
(j=2)
I
k
< 0.5 1400 1050
0.5 I
k
< 0.85 1600 1200
0.85 I
k
< 1.25 1700 1275
1.25 I
k
< 2.0 2000 1500
I
k
2.0 2300 1725

R = East South Central
Income Ratio:
Basement
No (j=1) Yes
2
(j=2)
I
k
< 0.5 1300 975
0.5 I
k
< 0.85 1400 1050
0.85 I
k
< 1.25 1700 1275
1.25 I
k
< 2.0 1900 1425
I
k
2.0 2500 1875

R = West South Central
Income Ratio:
Basement
No (j=1) Yes
2
(j=2)
I
k
< 0.5 1300 975
0.5 I
k
< 0.85 1700 1275
0.85 I
k
< 1.25 1800 1350
1.25 I
k
< 2.0 1900 1425
I
k
2.0 2500 1875
311
HazusMHTechnicalManual
Table 3.3 Typical Square Footage Per Unit (Main Living Area) by
Census Division (R)
1
(Continued)
R = Mountain
Income Ratio:
Basement
No (j=1) Yes
2
(j=2)
I
k
< 0.5 1200 900
0.5 I
k
< 0.85 1500 1125
0.85 I
k
< 1.25 1700 1275
1.25 I
k
< 2.0 1800 1350
I
k
2.0 2600 1950

R = Pacific
Income Ratio:
Basement
No (j=1) Yes
2
(j=2)
I
k
< 0.5 1300 975
0.5 I
k
< 0.85 1500 1125
0.85 I
k
< 1.25 1700 1275
1.25 I
k
< 2.0 1900 1425
I
k
2.0 2100 1575
Notes:
1 Based on data from the Energy Information Administration, Housing Characteristics 1993;
2 (Area of main living area if basement present) = 0.75 x (Area of main living area if no
basement). This adjustment allows consistent application of the Means cost models, in
which basement areas are added-on, and are assumed to be 1/3 of main living area.

While the US Census data does have data defining the median income for each census
block, there is data for the median income for each census block group. This value
will be applied to each block within the block group. With the a median income for
each census block, and the median income for the census region, it is possible to
define an Income Ratio that can be used to determine the square footage for buildings
with and without basements. Table 3.4 below shows the 9 census regions, the states
within those regions and the values used to compute the Income Ratio. The value
from the Census STF3A field P08A001 is the median income for the census block
group that will be applied to every census block within the group. The distribution of
basements is a summation or roll-up of the foundation type distribution discussed
later in this section.
312
Chapter3InventoryData:CollectionandClassification
Table 3.4 Income Ratio and Basement Distribution by Census Region


Region (States) Income Ratio
Percent with
Basement
Percent
without
Basement
AL P053001 / 36,268 25 75
AK P053001 / 52,492 13 87
AZ P053001 / 39,653 32 68
AR P053001 / 30,082 5 95
CA P053001 / 45,070 13 87
CO P053001 / 49,216 32 68
CT P053001 / 50,647 81 19
DE P053001 / 47,438 23 77
DC P053001 / 38,005 23 77
FL P053001 / 37,305 23 77
GA P053001 / 41,481 23 77
HI P053001 / 45,657 13 87
ID P053001 / 37,760 32 68
IL P053001 / 46,649 68 32
IN P053001 / 41,315 68 32
IA P053001 / 41,560 75 25
KS P053001 / 38,393 75 25
KY P053001 / 36,826 25 75
LA P053001 / 32,500 5 95
ME P053001 / 39,815 81 19
MD P053001 / 52,846 23 77
MA P053001 / 45,769 81 19
MI P053001 / 46,034 68 32
MN P053001 / 50,088 75 25
MS P053001 / 31,963 25 75
MO P053001 / 44,247 75 25
MT P053001 / 32,553 32 68
NE P053001 / 39,029 75 25
NV P053001 / 43,262 32 68
NH P053001 / 48,029 81 19
NJ P053001 / 51,739 76 24
NM P053001 / 34,035 32 68
NY P053001 / 40,822 76 24
NC P053001 / 38,413 23 77
ND P053001 / 33,769 75 25
313
HazusMHTechnicalManual
Table 3.4 Income Ratio and Basement Distribution by Census Region
(Continued)
Region (States) Income Ratio
Percent with
Basement
Percent
without
Basement
OH P053001 / 41,972 68 32
OK P053001 / 34,020 5 95
OR P053001 / 41,915 13 87
PA P053001 / 41,394 76 24
RI P053001 / 43,428 81 19
SC P053001 / 36,671 23 77
SD P053001 / 35,986 75 25
TN P053001 / 35,874 25 75
TX P053001 / 39,296 5 95
UT P053001 / 46,539 32 68
VT P053001 / 40,908 81 19
VA P053001 / 47,701 23 77
WA P053001 / 46,412 13 87
WV P053001 / 29,217 23 77
WI P053001 / 45,441 68 32
WY P053001 / 38,291 32 68

Once the parameters above had been defined, it is possible to develop an algorithm
that allows for the estimation of the RES1 or single-family residential square footage
for the entire nation. This algorithm is:

RES1 (sq. ft.) = Total Single Family Units (STF1B
H1BX0002) *[(Percent of units with basement) * (floor area
w/basement based on income ratio and region) + (Percent of
units without basement)*(floor area w/o basement based on
income ratio and region]

where Income Ratio = STF3A P08A001/regional income

For a sample New England census block, 81% Basement 19% no basement and an I
k

of 0.67:
RES1 (sq. ft.) = [STF1BX0002] * [(0.81)*(1,125) + (0.19)*(1,500)]

Multi-Family and Manufactured Housing (RES3 and RES2)

Developing the multi-family (RES3A through RES3F) and manufactured housing
(RES2) inventory requires additional information and effort compared to the single-
family occupancy classification. In the 1999 census extract, the STF1B (census block
314
Chapter3InventoryData:CollectionandClassification
data) extract identifies only those housing units within the 10 or more unit
classification, unfortunately, the 2000 census extract no longer provided that
information. Therefore in order to define of the multi-family units, it is necessary to
utilize the STF3A extract. The multi-family definition in the STF3A extract
identifies Duplex, 3-4 Unit, 5-9 unit, 10-19 unit, 20-49 unit, and 50+ dwellings.
Additionally the STF3A census data provides a definition of the Manufactured
Housing (MH) units within a block group and therefore the RES2 was processed at
the same time. The census data has an other classification for that will be ignored
since this classification represent a very small portion of the universe of housing units
and there is no other damage functions that can be assigned to these facilities.
Examples of the Other Census classification include vans and houseboats.

Unlike the single family residential that used the Housing Characteristics 1993 to
define heated floor area, assessor data from around the United States, including that
from the six Proof-of-Concept (POC) communities, was reviewed to develop
preliminary estimates of average floor area for multi-family housing. This data was
then peer reviewed by engineering experts to develop an average floor area per
number of units for the unit ranges provided by the census data. Table 3.5 shows the
distribution of the floor area by unit. The associated equations provide an example of
the calculations that have taken place.

Table 3.5 Floor Areas for Multi-Family Dwellings (RES2 & RES3A-RES3F)


Units Duplex 3-4 5-9 10-19 20-49 50+
Manufactured
Housing
Other
Floor
Area
1,500 750 800 750 700 650
Single Wide 950
Double Wide
1,350
NA

Previously, the flood model team had a complex process that allowed for a more
accurate block level distribution. However, when the US Census Bureau modified
the SF1 extract to eliminate information regarding the single-family and large multi-
family fields, it became necessary to modify the data manipulation process. The
multi-family data was still available in the SF3 extract at the census block group
level. The only available process was to distribute the census block group data
homogeneously throughout the census blocks. The distribution process is facilitated
by finding the ratio of total housing units per census block (H1BX0001) with respect
to the total housing units per census block group (H0010001). This ratio was then
used to as a multiplier to distribute the census block group level multi-family data
into each census block.

315
HazusMHTechnicalManual
Step 1: Develop the ratio of total housing units for each census block
Unit Ratio = (H1BX0001)/(H0010001)

Step 2: Distribute the multi-family housing units throughout each census block

For example:

Duplex units per block = H0200003*Unit Ratio

Step 3: Derive Floor area per occupancy classification

Manufactured Housing (sq. ft.) = Census Block RES2 (from Step 2)*
(0.75 * 950 + 0.25 * 1,400)
Duplex (sq. ft.) = (Census Block Duplex from Step 2) * 1,500

3-4 Units (sq. ft.) = (Census Block 3-4 units from Step 2) * 750

5-9 Units (sq. ft.) = (Census Block 5-9 units from Step 2) * 800

10-19 Units (sq. ft.) = (Census Block 10-19 units from Step 2) * 750

20-49 Units (sq. ft.) = (Census Block 20-49 units from Step 2) * 700

50+ Units (sq. ft.) = (Census Block 50+ units from Step 2) * 700

By using the above distribution, the valuation can be more specifically tailored to
each floor plan. This has the potential future benefit of allowing the user to modify
the floor area for multi-family units. For example in future releases, it may be
possible to provide the user the capability to modify the average floor area for
duplexes to 2,000 Sq Ft per unit if this more closely reflected the users community.
This should then lead to a net decrease in the total number of units for the RES3A
occupancy classification.
The floor areas presented for manufactured housing are based on review of various
internet websites for manufactured housing sales (new and used), housing
manufacturers, and finally additional US Census Bureau data. There was a great deal
of information regarding sales and shipment of manufactured housing since the
1970s, but there was very little information regarding the attrition rate experienced
over the same 30-year span. Charting information from the Manufactured Housing
Institute, Figure 3.1 shows that there has been a general growth trend in the size of the
units since the 1980s for both the single wide and doublewide (also known as single-
section and multi-section) manufactured housing.
316
Chapter3InventoryData:CollectionandClassification


Figure 3.1 Manufactured Housing Growth Over Time

The recently released American Housing Survey for the US, 1997
3
(September,
1999) contained estimated floor areas for manufactured housing (labeled Mobile
Home in the Census tables) based on a surveyed population of over 8 million
manufactured homes across the United States. The survey does not differentiate
between single-section and multi-section units, but when the values are charted the
distribution presents natural points to estimate these dimensions. Figure 3.2 shows
the distribution of floor area by number of structures from the survey. Using this
distribution, it is possible to estimate representative values for single-section and
multi-section units of 950 Square Feet and 1,400 Square Feet respectively.

3
US Department of Housing and Urban Development and US Census Bureau, American Housing
Survey for the United States H150/97, Office of Policy Development an Research and the US Census
Bureau, September 1999.
0
200
400
600
800
1000
1200
1400
1600
1800
1
9
8
0
1
9
8
1
1
9
8
2
1
9
8
3
1
9
8
4
1
9
8
5
1
9
8
6
1
9
8
7
1
9
8
8
1
9
8
9
1
9
9
0
1
9
9
1
1
9
9
2
1
9
9
3
1
9
9
4
1
9
9
5
1
9
9
6
1
9
9
7
Year of Sal e
A
v
e
r
a
g
e

F
l
o
o
r

A
r
e
a
Single Wide
Double Wide
317
HazusMHTechnicalManual

Figure 3.2 Number of Mobile Home Units by Floor Area 1990 US Census Data
Housing Characteristics

Non-Residential Occupancy Classifications

The Hazus99 Earthquake Model inventory used the D&B business inventory at the
census tract level for all non-residential structures and those facilities that are
commercial in nature but provide housing for people such as hotels (RES4) and
nursing homes (RES6). The D&B data represents approximately 76 percent
(approximately 14 million) of the total estimated businesses in the United States
(approximately 19 million). While initially this might seem like a low representation,
the D&B database accounts for 98 percent of the gross national product. D&B states
that the remaining businesses are likely to be smaller and home-based. If true, the
proxy inventory established for the residential dwellings will account for these
businesses in the total damage estimates.

D&B provided the data aggregated on the SIC definitions used previously in the
development of the Hazus99 Earthquake Model (Hazus99 Users Manual, 1997 Table
Appendix A.19, page A-23). The D&B data obtained for the Flood Model provided
floor area for businesses at the census block level. It should be noted that D&B
performs regular random sampling of businesses in their database to obtain the actual
floor area. D&B then utilizes proprietary algorithms to estimate the floor area for the
remaining businesses. According to D&B, floor area is sampled for approximately 25
percent of their business database and the remainder is modeled.
0
500
1000
1500
2000
2500
0
2
0
0
4
0
0
6
0
0
8
0
0
1
0
0
0
1
2
0
0
1
4
0
0
1
6
0
0
1
8
0
0
2
0
0
0
2
2
0
0
2
4
0
0
2
6
0
0
2
8
0
0
3
0
0
0
3
2
0
0
3
4
0
0
3
6
0
0
3
8
0
0
4
0
0
0
4
2
0
0
4
4
0
0
Floor Area (Square Feet)
N
u
m
b
e
r

o
f

M
o
b
i
l
e

H
o
m
e

U
n
i
t
s

(
T
h
o
u
s
a
n
d
s
)
US Census Total Units Surveyed = 8,301,000
318
Chapter3InventoryData:CollectionandClassification

With their data, D&B provided a count of businesses, the total floor area (modeled
and sampled), and the total number of employees. During a review of the data, it was
discovered that D&B had some data aggregated at the census block groups and tracts
level. Review of the data determined that these errors were consistent with automated
georeferencing processes and are likely to represent those businesses where the
addresses did not match directly with D&Bs reference street base. D&B performed
an additional review and ascertained that this was in fact the cause of this
aggregation. It was felt, however, that the tract and block group data could be safely
distributed to the census blocks based on weighted averages of commercial
development within the blocks. Review of the results of this effort showed little net
impact and continued agreement with ground truth data.

The D&B data contained information on all non-residential uses including some
agricultural facilities, general government offices, schools, and churches. Again,
comparison with POC data and other available data showed relatively good
agreement.

3.2.1.5. Building Replacement Costs

Building replacement cost models within Hazus are based on industry-standard cost-
estimation models published in Means Square Foot Costs (R.S. Means, 2002).
Replacement cost data are stored within Hazus at the census tract and census block
level for each occupancy class. For each Hazus occupancy class, a basic default
structure full replacement cost model (cost per square foot) has been determined, and
are provided in Table 3.6. Commercial and industrial occupancies have a typical
building replacement cost model associated with each occupancy class (e.g., COM4,
Professional/Technical/Business Services, is represented by a typical, 80,000 square
foot, 5 to 10 story office building). In most cases, the typical building chosen to
represent the occupancy class is the same as was used in the original Hazus
earthquake model (based on Means, 1994), except for single family residential, multi-
family residential, and industrial uses. Both primary default (in bold) and alternate
models (in italics) are provided in the table. As shown, in some cases the alternate
costs are very similar to the primary default cost (e.g., medium and large
dormitories), although a number of alternate costs vary quite a bit (medium hotel vs.
medium motel). Square foot costs presented in the table have been averaged over the
various alternatives for exterior wall construction (e.g., wood siding over wood
frame, brick veneer over wood frame, stucco on wood frame or precast concrete,
concrete block over wood joists or precast concrete, etc.). For non-residential
structures, the default configuration assumes structures without basements.

The RES1 (single family residential) replacement cost model is the most complex,
utilizing socio-economic data from the census to determine an appropriate mix of
construction classes (Economy, Average, Custom and Luxury) and associated
319
HazusMHTechnicalManual
replacement cost models. The algorithm is described in Section 3.2.1.5.1.4. Within
Means, basements are not considered in the base cost of the structure and are handled
as an additive adjustment (additional cost per square foot of main structure). Table
3.7 provides Means (2002) replacement costs for the various single family dwelling
configurations available in the default building inventory (1, 2, and 3 story and split-
level), assuming a typical size of 1,600 square feet. Costs have been averaged for the
various alternatives for exterior wall construction.

Because the default single family residential (SFR) damage model is based on the
FIA credibility-weighted depth damage functions, whose coverage extends to
garages, the replacement cost of garages will also be included in the basic
replacement cost. Relevant Means models for SFR garages include costs by
construction class (economy, average, custom, and luxury), for detached and attached
1-car, 2-car and 3-car garages, constructed of wood or masonry. For incorporation
into Hazus, costs by size and construction class have been averaged for
attached/detached and various materials. Average costs associated with garage types
included in the default inventory for single family residential structures (1-car, 2-car
and 3-car) were provided in Table 3.8.

3.2.1.5.1. Single-Family Residential Valuation Algorithm

The algorithm defined below will be used to develop the valuation for single-family
residential buildings at the census block level. This algorithm utilizes socio-
economic data from the census to derive an appropriate Means-based cost for each
census block. The earthquake and wind models shall use a roll-up of the results
from the flood model calculations. Some round-off error will occur, but this cannot
be avoided.

The valuation algorithm can be summarized mathematically in equation (1) below:


Where:

V
RES1, k
is the total estimated valuation for single-family
residences (RES1) for a given census block (k). V
RES1, k

is editable when viewing the dollar exposure by specific
occupancy table.


4 4

4 4

V
RES1, k
= (A
RES1
,
k
)*[ w
i,k
*w
j,k
*C
i,j
] + (A
RES1
,
k
)*w
l,k
*[ ?w
i,k
*w
j,k
*C
i,j,l
]
i=1 j=1 i=1 j=1

4

4

+(RES1Cnt
k
)*[ ? w
i,k
*w
m,k
*C
i, m
] (3-1)

i=1 j=1

320
Chapter3InventoryData:CollectionandClassification
A
RES1,k
is the total single-family residential (RES1) floor area
(square feet) for a given census block (k) found in the
square foot by specific occupancy table. A
RES1,k
is
editable when viewing the square foot by specific
occupancy table.

i the Means construction class (1 = Economy, 2 =
Average,
3 = Custom, 4 = Luxury).

w
i,k
is the weighting factor for the Means construction class
(i) for the given census block (k) and is determined
from the income ratio range as shown in Table 14.4
below. Values are displayed in percent to the user and
are editable when viewing the dollar exposure
parameters tables.

j the number of stories class for single-family (RES1)
structures
(1 = 1-story, 2 = 2-story, 3 = 3-story, and 4 = split
level)

w
j,k
is the weighting factor for the Number of Stories class
(j) for the given census block (k) depending on the
census region of that block (by state FIPS). Weighting
factors were developed from regional construction type
distributions as discussed in Section 3. Values are
displayed in percent to the user and are editable when
viewing the dollar exposure parameters tables.

C
i,j
is the single-family (RES1) cost per square foot for the
given Means construction class (i) and number of
stories class (j). RES1 replacement costs are seen in the
third column of Table 14.2. Values are editable when
viewing the dollar exposure parameters tables.

l the basement status available for single-family
residences
(1 = yes, 2 = no).

w
l,k
is the weighting factor for basements for the given
census block (k) depending on the census region of that
block (by state FIPS). Weighting factors were
developed from regional foundation type distributions
as discussed in Section 3. Values are displayed in
321
HazusMHTechnicalManual
percent to the user and are editable when viewing the
dollar exposure parameters tables. Default will be
established based on whether the block is a coastal or
non-coastal block.

C
i,j,l
the additional cost, per square foot of the main
structure, for a finished basement for the given Means
construction class (i) and number of stories class (j), as
shown in Table 14.2, Column 4. Note: C
i,j,l
= 0 when l
= 2. Values are editable when viewing the dollar
exposure parameters tables.
m the garage combinations available for single-family
residences (1 = 1-car, 2 = 2-car, 3 = 3-car, 4 = carport,
and 5 = none).

w
m,k
is the weighting factor for the garage type (m) for the
given census block (k) depending on the census region
of that block (by state FIPS). Weighting factors were
developed from regional construction type distributions
as discussed in Section 3. Values are displayed in
percent to the user and are editable when viewing the
dollar exposure parameters tables.

C
i,m
the additional replacement cost for a given garage type
(m), for the given Means construction class (i) as shown
in Table 14.3. Note: C
i,m
= 0 when m = 4 (covered
carport) or m = 5 (none). Values are editable when
viewing the dollar exposure parameters table.

R
ES1Cnt
the count of RES1 structures within the given census
block (k) taken directly from the Building Count by
occupancy table.

As the algorithm shows, the basic replacement cost per square foot is a function of the
Means construction class, the number of stories and an additional cost per square foot
of the main structure for the existence of a finished or unfinished basement. Finally,
there is an additional cost per housing unit based on the garage associated with the
structure. The valuation parameters are presented in a series of tables in Section
3.2.1.5.1.4 of this document.

3.2.1.5.2. Manufactured Housing Valuation Algorithm

It is necessary to clarify that RES2 within Hazus99 and Hazus
MH
, while designated
Manufactured Housing, represents Mobile Homes and not single-family pre-
322
Chapter3InventoryData:CollectionandClassification
manufactured housing. The US Census provides a detailed count of the mobile
homes within each census block and this quantity is used to develop the total floor
area (square foot) of the RES2 occupancy classification. The total floor area was
developed assuming a typical floor area and average distribution of singlewide to
doublewide mobile homes. Unlike other occupancy classifications, there are no
allowances for variation of floor heights (number of stories) or other valuation
parameters. The valuation of manufactured housing is the straight multiplication of
the total floor area by the baseline replacement cost per square foot. The cost per
square foot (C
RES2
) is defined in Table 3.6 in the valuation parameters section
(Section 3.2.1.5.1.4) of this document

The algorithm for manufactured housing is defined in equation (2) below:

V
RES2,k
= A
RES2,k
*C
RES2
(3-2)

Where:

V
RES2,k
is the total estimated valuation for Manufactured
Housing (RES2) for a given census block (k). V
RES2
,
k

is editable when viewing the dollar exposure by specific
occupancy table.

A
RES2,k
is the total Manufactured Housing (RES2) floor area
(square feet) for a given census block (k) found in the
square foot by specific occupancy table. A
RES2,k
is
editable when viewing the square foot by specific
occupancy table.

C
RES2
is the Manufactured Housing (RES2) cost per square
foot. RES2 replacement costs are Table 14.1
($30.90/SqFt). The value is editable when viewing the
dollar exposure parameters tables.

The flood model has accounted for differential areas between singlewide and
doublewide manufactured housing in the total floor area, it is assumed that the cost
per square foot does not vary greatly between the two structure types.

3.2.1.5.3. Other Residential and Non-Residential Occupancies

The algorithm for the remaining residential occupancies (RES3-RES6) and all non-
residential (COM, IND, EDU, REL, GOV, and AGR) occupancies is not as complex
as the single family model but allows for the potential incorporation of a distribution
323
HazusMHTechnicalManual
for number of stories. It should be noted that the replacement costs seen in Table 3.6
are an average replacement cost by occupancy. In other words, the replacement cost
is averaged across structure types, stories and construction classes to produce the
values in Table 14.1.

The algorithm for the remaining residential occupancies and non-residential
occupancies can be seen in equation (3) below:

V
x,k
= A
x,k
*C
x
(3-3)

Where:

x defines the remaining occupancy classifications (x
ranges from 3 to 28 for the remaining occupancies, i.e.,
RES5, COM1, REL1, etc.) for which the cost is being
calculated.

V
x,k
is the total estimated valuation for the specific
occupancy (x) (such as RES4, COM3, or IND6) for a
given census block (k). V
x
,
k
is editable when viewing
the dollar exposure by specific occupancy table.

A
x,k
is the total floor area (square feet) for a specific
occupancy (x) (such as RES3, COM8, IND4, GOV1,
etc.) for a given census block (k) found in the square
foot by specific occupancy table. A
x,k
is editable when
viewing the square foot by specific occupancy table.

C
x
is the cost per square foot for the specific occupancy
(x). The replacement costs are seen in Table 14.1
below by specific occupancy. Values are editable when
viewing the dollar exposure parameters tables.

At this time, the flood model depreciation models for non-single-family residential
structures will not depend on features such as the number of stories. A distribution of
number of stories will still be developed in the dollar exposure parameters table since
the creation of such depreciation models are seen as a potential enhancement in future
versions of the Hazus Flood Model.

3.2.1.5.4. Valuation Tables

324
Chapter3InventoryData:CollectionandClassification
The following tables present the baseline valuation parameters for the variables
discussed in Section 14.4 of this document. Each of these parameters is editable by
the user.

Table 3.6 Default Full Replacement Cost Models (Means, 2002)

Hazus Occupancy Class
Description
Sub-category
Means Model Description (Means
Model Number)
Means
Cost/SF
(2002)
RES1
Single Family
Dwelling
See Table 14.2
RES2
Manufactured
Housing
Manufactured
Housing
Manufactured Housing (N/A)
1

$30.90
RES3
Multi Family
Dwelling small
Duplex SFR Avg 2 St., MF adj, 3000 SF $67.24
Triplex/Quads SFR Avg 2 St., MF adj, 3000 SF $73.08
Multi Family
Dwelling medium
5-9 units Apt, 1-3 st, 8,000 SF (M.010) $125.63
10-19 units Apt., 1-3 st., 12,000 SF (M.010) $112.73
Multi Family
Dwelling large
20-49 units Apt., 4-7 st., 40,000 SF (M.020) $108.86
50+ units Apt., 4-7 st., 60,000 SF (M.020) $106.13
Apt., 8-24 st., 145,000 SF (M.030) $111.69
RES4 Temp. Lodging
Hotel, medium Hotel, 4-7 st., 135,000 SF(M.350) $104.63
Hotel, large Hotel, 8-24 st., 450,000 SF (M.360) $93.47
Motel, small Motel, 1 st., 8,000 SF (M.420) $94.13
Motel, medium Motel, 2-3 st., 49,000 SF (M.430) $110.03
RES5
Institutional
Dormitory
Dorm, medium
College Dorm, 2-3 st, 25,000 SF
(M.130)
$118.82
Dorm, large
College Dorm, 4-8 st, 85,000 SF
(M.140)
$113.31
Dorm, small Frat House, 2 st., 10,000 SF (M.240) $99.50
RES6 Nursing Home Nursing home
Nursing Home, 2 st., 25,000 SF
(M.450)
$104.62
COM1 Retail Trade
Dept Store, 1 st Store, Dept., 1 st., 110,000 SF (M.610) $71.54
Dept Store, 3 st Store, Dept., 3 st., 95,000 SF (M.620) $88.73
Store, small Store, retail, 8,000 SF (M.630) $79.23
Store, medium Supermarket, 44,000 SF (M.640) $69.09
Store, convenience Store, Convenience, 4,000 SF (M.600) $83.59
Auto Sales
Garage, Auto Sales, 21,000 SF
(M.260)
$70.84
COM2 Wholesale Trade
Warehouse, medium Warehouse, 30,000 SF (M.690) $61.91
Warehouse, large Warehouse, 60,000 SF (M.690) $56.58
Warehouse, small Warehouse, 15,000 SF (M.690) $70.43
COM3
Personal and Repair
Services
Garage, Repair Garage, Repair, 10,000 SF (M.290) $86.81
Garage, Service sta. Garage, Service sta., 1,400 SF (M.300) $113.91
Funeral Home Funeral home, 10,000 SF (M.250) $97.66
Laundromat Laundromat 3,000 SF (M.380) $135.64
Car Wash Car Wash, 1 st., 800 SF (M.080) $198.28
325
HazusMHTechnicalManual
COM4
Prof./
Tech./Business
Services
Office, Medium Office, 5-10 st., 80,000 SF (M.470) $98.96
Office, Small Office, 2-4 st., 20,000 SF (M.460) $102.69
Office, Large Office, 11-20 st., 260,000 SF (M.480) $88.21
COM5 Banks Bank Bank, 1 st., 4100 SF (M.050) $153.97
326
Chapter3InventoryData:CollectionandClassification
Table 3.6 Default Full Replacement Cost Models (Means, 2002) (Continued)
Hazus Occupancy Class
Description
Sub-category
Means Model Description (Means
Model Number)
Means
Cost/SF
(2002)
COM6 Hospital
Hospital, Medium Hospital, 2-3 st., 55,000 SF (M.330) $144.60
Hospital, Large Hospital, 4-8 st., 200,000 SF (M.340) $125.60
COM7 Medical Office/Clinic
Med. Office, medium Medical office, 2 st., 7,000 SF (M.410) $129.82
Med. Office, small Medical office, 1 st., 7,000 SF (M.400) $118.01
COM8
Entertainment &
Recreation
Restaurant Restaurant, 1 st., 5,000 SF (M.530) $137.02
Restaurant, Fast food
Restaurant, fast food, 4,000 SF
(M.540)
$121.49
Bowling Alley Bowling Alley, 20,000 SF (M.060) $72.31
Country Club Club, Country, 1 st., 6,000 SF (M.100) $135.23
Social Club Club, Social, 1 st., 22,000 SF (M.110) $95.39
Racquetball Court Racquetball Court, 30,000 SF (M.510) $111.23
Hockey Rink Hockey Rink 30,000 SF (M.550) $115.13
COM9 Theaters
Movie Theatre Movie Theatre, 12,000 SF (M.440) $102.35
Auditorium Auditorium, 1 st., 24,000 SF (M.040) $109.60
COM10 Parking
Parking Garage
Garage, Pkg, 5 st., 145,000 SF
(M.270)
$34.78
Parking Garage,
Underground
Garage, UG Pkg, 100,000 SF (M.280) $49.20
IND1 Heavy
Factory, small Factory, 1 st., 30,000 SF (M.200) $73.82
Factory, large Factory, 3 st., 90,000 SF (M.210) $78.61
IND2 Light
Warehouse, medium Warehouse, 30,000 SF (M.690) $61.91
Factory, small Factory, 1 st., 30,000 SF (M.200) $73.82
Factory, large Factory, 3 st., 90,000 SF (M.210) $78.61
IND3 Food/Drugs/Chemicals
College Laboratory College Lab, 1 st., 45,000 SF (M.150) $119.51
Factory, small Factory, 1 st., 30,000 SF (M.200) $73.82
Factory, large Factory, 3 st., 90,000 SF (M.210) $78.61
IND4
Metals/Minerals
Processing
College Laboratory College Lab, 1 st., 45,000 SF (M.150) $119.51
Factory, small Factory, 1 st., 30,000 SF (M.200) $73.82
Factory, large Factory, 3 st., 90,000 SF (M.210) $78.61
IND5 High Technology
College Laboratory College Lab, 1 st., 45,000 SF (M.150) $119.51
Factory, small Factory, 1 st., 30,000 SF (M.200) $73.82
Factory, large Factory, 3 st., 90,000 SF (M.210) $78.61
IND6 Construction
Warehouse, medium Warehouse, 30,000 SF (M.690) $61.91
Warehouse, large Warehouse, 60,000 SF (M.690) $56.58
Warehouse, small Warehouse, 15,000 SF (M.690) $70.43
AGR1 Agriculture
Warehouse, medium Warehouse, 30,000 SF (M.690) $61.91
Warehouse, large Warehouse, 60,000 SF (M.690) $56.58
Warehouse, small Warehouse, 15,000 SF (M.690) $70.43
REL1 Church Church Church, 1 st., 17,000 SF (M.090) $114.08
327
HazusMHTechnicalManual
Table 3.6 Default Full Replacement Cost Models (Means, 2002) (Continued)
Hazus Occupancy Class
Description
Sub-category
Means Model Description (Means
Model Number)
Means
Cost/SF
(2002)
GOV1 General Services
Town Hall, small Town Hall, 1 st., 11,000 SF (M.670) $90.30
Town Hall, medium
Town Hall, 2-3 st., 18,000 SF
(M.680)
$112.94
Courthouse, small Courthouse, 1 st., 30,000 SF (M.180) $130.71
Courthouse, medium
Courthouse, 2-3 st., 60,000 SF
(M.190)
$136.81
Post Office Post Office, 13,000 SF (M.500) $86.83
GOV2 Emergency Response
Police Station
Police Station, 2 st., 11,000 SF
(M.490)
$136.10
Fire Station, small Fire Station, 1 st., 6,000 SF (M.220) $105.53
Fire Station, medium Fire Station, 2 st., 10,000 SF (M.230) $110.34
EDU1 Schools/Libraries
High School School, High, 130,000 SF (M.570) $92.80
Elementary School
School, Elementary, 45,000 SF
(M.560)
$90.22
Jr. High School School, Jr. High, 110,000 SF (M.580) $95.21
Library Library, 2 st., 22,000 SF (M.390) $103.94
Religious School
Religious Educ, 1 st., 10,000 SF
(M.520)
$112.19
EDU2 Colleges/Universities
College Classroom
College Class. 2-3 st, 50,000 SF
(M.120)
$114.68
College Laboratory College Lab, 1 st., 45,000 SF (M.150) $119.51
Vocational school
School, Vocational, 40,000 SF
(M.590)
$93.96
Notes:
1 Manufactured Housing Institute, 2000 cost for new manufactured home

328
Chapter3InventoryData:CollectionandClassification
Table 3.7 Replacement Costs (and Basement Adjustment) for RES1 Structures
by Means Constructions Class (Means, 2002)

Means
Construction
Class
Height
Class
Average Base
cost per
square foot
Adjustment for
Finished Basement
(cost per SF of
main str.)
Adjustment for
Unfinished
Basement
(cost per SF of
main str.)
Economy
1 story 55.23 16.95 6.35
2 story 59.58 9.85 4.20
3-story N/A use 2 st N/A use 2 st N/A use 2 st
Split level* 55.30 12.32 5.02
Average
1 story 79.88 21.15 7.35
2 story 79.29 13.80 4.85
3-story 84.81 10.97 3.78
Split level 74.94 16.42 5.77
Custom
1 story 99.59 31.90 11.65
2 story 99.63 18.75 7.15
3-story 105.83 13.78 5.35
Split level 93.81 23.35 8.78
Luxury
1 story 122.25 37.75 13.20
2 story 117.55 22.35 8.10
3-story 124.00 16.48 6.05
Split level 111.13 27.82 9.95

Table 3.8 Single Family Residential Garage Adjustment (Means, 2002)

Means Construction Class Garage Type
Average Additional Garage Cost
per Residence
Economy
1 car $10,700
2 car $16,700
3 car $22,600
Average
1 car $11,000
2 car $17,100
3 car $23,000
Custom
1 car $12,500
2 car $19,700
3 car $26,600
Luxury
1 car $14,700
2 car $23,300
3 car $31,700

329
HazusMHTechnicalManual
Table 3.8 Weights (percent) for Means Construction/Condition Models

Income
Weights (w) for:
c
Lg
c
Cg
c
Aa
c
Ep

I
k
< 0.5 - - 70 30
0.5 I
k
< 0.85 - - 100 -
0.85 I
k
< 1.25 - - 100 -
1.25 I
k
< 2.0 - 60 40 -
I
k
2.0 40 60 - -

3.2.1.6. Contents Replacement Cost

Contents replacement value is estimated as a percent of structure replacement value.
The NIBS Flood Module will utilize the same contents to structure value ratios as are
employed in the NIBS Earthquake Module (Table 15.5 in the Hazus 1999 Technical
Manual), provided in Table 3.10.

Table 3.10 Default Hazus Contents Value Percent of Structure Value

No. Label Occupancy Class Contents Value (%)
Residential
1 RES1 Single Family Dwelling 50
2 RES2 Mobile Home 50
3 RES3 Multi Family Dwelling 50
4 RES4 Temporary Lodging 50
5 RES5 Institutional Dormitory 50
6 RES6 Nursing Home 50
Commercial
7 COM1 Retail Trade 100
8 COM2 Wholesale Trade 100
9 COM3 Personal and Repair Services 100
10 COM4 Professional/Technical/
Business Services
100
11 COM5 Banks 100
12 COM6 Hospital 150
13 COM7 Medical Office/Clinic 150
14 COM8 Entertainment & Recreation 100
15 COM9 Theaters 100
16 COM10 Parking 50
Industrial
17 IND1 Heavy 150
18 IND2 Light 150
19 IND3 Food/Drugs/Chemicals 150
20 IND4 Metals/Minerals Processing 150
330
Chapter3InventoryData:CollectionandClassification
No. Label Occupancy Class Contents Value (%)
21 IND5 High Technology 150
22 IND6 Construction 100
Agriculture
23 AGR1 Agriculture 100
Religion/Non/Profit
24 REL1 Church/Membership Organization 100
Government
25 GOV1 General Services 100
26 GOV2 Emergency Response 150
Education
27 EDU1 Schools/Libraries 100
28 EDU2 Colleges/Universities 150

3.2.2. Essential Facilities

Essential facilities are those facilities that provide services to the community and
should be functional after an earthquake. Essential facilities include hospitals, police
stations, fire stations and schools. The damage state probabilities for essential
facilities are determined on a site-specific basis (i.e., the ground motion parameters
are computed at the location of the facility). The purpose of the essential facility
module is to determine the expected loss of functionality for these critical facilities.
Economic losses associated with these facilities are computed as part of the analysis
of the general building stock (general building stock occupancy classes 12, 26, 27 and
28). The data required for the analysis include mapping of essential facilitys
occupancy classes to model building types or a combination of essential facilities
building type, design level and construction quality factor. In addition, the number of
beds for each hospital and the number of fire trucks at each fire station are required.
The fire truck information is used as input for the fire following earthquake analysis
(Chapter 10).

3.2.2.1. Classification

The essential facilities are also classified based on the building structure type and
occupancy class. The building structure types of essential facilities are the same as
those for the general building stock presented in Table 3.1. The occupancy
classification is broken into general occupancy and specific occupancy classes. For
the methodology, the general occupancy classification system consists of three groups
(medical care, emergency response, and schools). Specific occupancy consists of nine
classes. The occupancy classes are given in Table 3.11, where the general occupancy
classes are identified in boldface. Relationships between specific and general
occupancy classes are in a form shown in Table 3B.1 of Appendix 3B.

331
HazusMHTechnicalManual

Table 3.11: Essential Facilities Classification

Label Occupancy Class Description
Medical Care Facilities
EFHS Small Hospital Hospital with less than 50 Beds
EFHM Medium Hospital Hospital with beds between 50 & 150
EFHL Large Hospital Hospital with greater than 150 Beds
EFMC Medical Clinics Clinics, Labs, Blood Banks
Emergency Response
EFFS Fire Station
EFPS Police Station
EFEO Emergency Operation Centers
Schools
EFS1 Grade Schools Primary/ High Schools
EFS2 Colleges/Universities


3.2.2.2. Occupancy to Model Building Type Relationship

Default mapping of essential facility occupancy classes to model building types is
provided in Tables 3B.2 through 3B.16 of Appendix 3B. For the regional designation
of a particular state, refer to Table 3C.1 in Appendix C. The default mapping of
specific occupancy to model building type mapping is based on general building
stock occupancy classes 12, 26, 27 and 28.

3.2.3. High Potential Loss Facilities

High potential loss facilities are facilities that are likely to cause heavy earthquake
losses if damaged. For this methodology, high potential loss (HPL) facilities include
nuclear power plants, dams, and some military installations. The inventory data
required for HPL facilities include the geographical location (latitude and longitude)
of the facility. Damage and loss estimation calculation for high potential loss
facilities are not performed as part of the methodology.

332
Chapter3InventoryData:CollectionandClassification
3.2.3.1. Classification

Three types of HPL facilities are identified in the methodology (dams, nuclear power
facilities and military installations) are shown in Table 3.12. The dam classification
is based on the National Inventory of Dams (NATDAM) database (FEMA, 1993).

Table 3.12: High Potential Loss Facilities Classification
Label Description
Dams
HPDE Earth
HPDR Rock fill
HPDG Gravity
HPDB Buttress
HPDA Arch
HPDU Multi-Arch
HPDC Concrete
HPDM Masonry
HPDS Stone
HPDT Timber Crib
HPDZ Miscellaneous

Nuclear Power Facilities
HPNP Nuclear Power Facilities

Military Installations
HPMI Military Installations

3.3. Direct Damage Data - Transportation Systems

The inventory classification scheme for lifeline systems separates components that
make up the system into a set of pre-defined classes. The classification system used
in this methodology was developed to provide an ability to differentiate between
varying lifeline system components with substantially different damage and loss
characteristics. Transportation systems addressed in the methodology include
highways, railways, light rail, bus, ports, ferries and airports. The classification of
each of these transportation systems is discussed in detail in the following sections.
The inventory data required for the analysis of each system is also identified in the
following sections.

For some transportation facilities, classification of the facility is based on whether the
equipment is anchored or not. Anchored equipment in general refers to equipment
designed with special seismic tie-downs or tiebacks, while unanchored equipment
refers to equipment designed with no special considerations other than the
manufacturer's normal requirements. While some vibrating components, such as
pumps, are bolted down regardless of concern for earthquakes, as used here
anchored means all components have been engineered to meet seismic criteria
which may include bracing (e.g., pipe or stack bracing) or flexibility requirements
(e.g., flexible connections across separation joints) as well as anchorage.
333
HazusMHTechnicalManual
3.3.1. Highway Systems
A highway transportation system consists of roadways, bridges and tunnels. The
inventory data required for analysis include the geographical location, classification,
and replacement cost of the system components. The analysis also requires the length
of each highway segment.
3.3.1.1. Classification
The classes of highway system components are presented in Table 3.13. For more
details on how to classify these components, refer to section 7.1.5 of Chapter 7.

Table 3.13: Highway System Classification
Label Description
Hignhway Roads
HRD1 Major Roads
HRD2 Urban Roads
Highnway Bridges
HWB1 Major Bridge - Length > 150m (Conventional Design)
HWB2 Major Bridge - Length > 150m (Seismic Design)
HWB3 Single Span (Not HWB1 or HWB2) (Conventional Design)
HWB4 Single Span (Not HWB1 or HWB2) (Seismic Design)
HWB5 Concrete, Multi-Column Bent, Simple Support (Conventional Design), Non-California (Non-CA)
HWB6 Concrete, Multi-Column Bent, Simple Support (Conventional Design), California (CA)
HWB7 Concrete, Multi-Column Bent, Simple Support (Seismic Design)
HWB8 Continuous Concrete, Single Column, Box Girder (Conventional Design)
HWB9 Continuous Concrete, Single Column, Box Girder (Seismic Design)
HWB10 Continuous Concrete, (Not HWB8 or HWB9) (Conventional Design)
HWB11 Continuous Concrete, (Not HWB8 or HWB9) (Seismic Design)
HWB12 Steel, Multi-Column Bent, Simple Support (Conventional Design), Non-California (Non-CA)
HWB13 Steel, Multi-Column Bent, Simple Support (Conventional Design), California (CA)
HWB14 Steel, Multi-Column Bent, Simple Support (Seismic Design)
HWB15 Continuous Steel (Conventional Design)
HWB16 Continuous Steel (Seismic Design)
HWB17 PS Concrete Multi-Column Bent, Simple Support - (Conventional Design), Non-California
HWB18 PS Concrete, Multi-Column Bent, Simple Support (Conventional Design), California (CA)
HWB19 PS Concrete, Multi-Column Bent, Simple Support (Seismic Design)
HWB20 PS Concrete, Single Column, Box Girder (Conventional Design)
HWB21 PS Concrete, Single Column, Box Girder (Seismic Design)
HWB22 Continuous Concrete, (Not HWB20/HWB21) (Conventional Design)
HWB23 Continuous Concrete, (Not HWB20/HWB21) (Seismic Design)
HWB24 Same definition as HWB12 except that the bridge length is less than 20 meters
HWB25 Same definition as HWB13 except that the bridge length is less than 20 meters
HWB26 Same definition as HWB15 except that the bridge length is less than 20 meters and Non-CA
HWB27 Same definition as HWB15 except that the bridge length is less than 20 meters and in CA
HWB28 All other bridges that are not classified (including wooden bridges)

Highwnay Tunnels
HTU1
Highway Bored/Drilled Tunnel
HTU2
Highway Cut and Cover Tunnel

334
Chapter3InventoryData:CollectionandClassification
3.3.2. Railways
A railway transportation system consists of tracks, bridges, tunnels, stations, and fuel,
dispatch and maintenance facilities. The inventory data required for analysis include
the geographical location, classification and replacement cost of the facilities,
bridges, tunnels, and track segments. The analysis also requires the length of the
railway segments.
3.3.2.1.Classification
The various classes of railway system components are presented in Table 3.14. For
more details on how to classify these components refer to section 7.2 of Chapter 7.

Table 3.14: Railway System Classification
Label Description
Railway Tracks
RTR1 Railway Tracks
Railway Bridges
RLB1
Steel, Multi-Column Bent, Simple Support (Conventional Design), Non-
California (Non-CA)
RLB2
Steel, Multi-Column Bent, Simple Support (Conventional Design), California
(CA)
RLB3 Steel, Multi-Column Bent, Simple Support (Seismic Design)
RLB4 Continuous Steel (Conventional Design)
RLB5 Continuous Steel (Seismic Design)
RLB6 Same definition as HWB1 except that the bridge length is less than 20 meters
RLB7 Same definition as HWB2 except that the bridge length is less than 20 meters
RLB8
Same definition as HWB4 except that the bridge length is less than 20 meters
and Non-CA
RLB9
Same definition as HWB5 except that the bridge length is less than 20 meters
and in CA
RLB10 All other bridges that are not classified
Railway Urban Station
RST Rail Urban Station (with all building type options enabled)
Railway Tunnels
RTU1 Rail Bored/Drilled Tunnel
RTU2 Rail Cut and Cover Tunnel
Railway Fuel Facility
RFF Rail Fuel Facility (different combinations for with or without anchored
components and/or with or without backup power)
Railway Dispatch Facility
RDF Rail Dispatch Facility (different combinations for with or without anchored
components and/or with or without backup power)
Railway Maintenance Facility
RMF Rail Maintenance Facility (with all building type options enabled)
335
HazusMHTechnicalManual
3.3.3. Light Rail

Like railways, light rail systems are composed of tracks, bridges, tunnels, and
facilities. The major difference between the two is with regards to power supply,
where light rail systems operate with DC power substations. The inventory data
required for analysis include the classification, geographical location, and
replacement cost of facilities, bridges, tunnels, and tracks. In addition, the analysis
requires the track length.

3.3.3.1. Classification
Table 3.15 describes the various classes of light rail system components. For more
details on how to classify these components refer to section 7.3 of Chapter 7.

Table 3.15: Light Rail System Classification

Label Description
Light Rail Tracks
LTR1 Light Rail Track

Light Rail Bridges
LRB1
Steel, Multi-Column Bent, Simple Support (Conventional Design), Non-California
(Non-CA)
LRB2 Steel, Multi-Column Bent, Simple Support (Conventional Design), California (CA)
LRB3 Steel, Multi-Column Bent, Simple Support (Seismic Design)
LRB4 Continuous Steel (Conventional Design)
LRB5 Continuous Steel (Seismic Design)
LRB6 Same definition as HWB1 except that the bridge length is less than 20 meters
LRB7 Same definition as HWB2 except that the bridge length is less than 20 meters
LRB8
Same definition as HWB4 except that the bridge length is less than 20 meters and
Non-CA
LRB9
Same definition as HWB5 except that the bridge length is less than 20 meters and
in CA
LRB10 All other bridges that are not classified

Light Rail Tunnels
LTU1 Light Rail Bored/Drilled Tunnel
LTU2 Light Rail Cut and Cover Tunnel

DC Substation
LDC1 Light Rail DC Substation w/ Anchored Sub-Components
LDC2 Light Rail DC Substation w/ Unanchored Sub-Components

Dispatch Facility
LDF Light Rail Dispatch Facility (different combinations for with or without
anchored components and/or with or without backup power)

Maintenance Facility
LMF Maintenance Facility (with all building type options enabled)


336
Chapter3InventoryData:CollectionandClassification
3.3.4. Bus System

A bus transportation system consists of urban stations, fuel facilities, dispatch
facilities and maintenance facilities. The inventory data required for bus systems
analysis include the geographical location, classification, and replacement cost of bus
system facilities.

3.3.4.1. Classification

Table 3.16 describes the various classes of bus system components. For more details
on how to classify these components refer to section 7.4 of Chapter 7.

Table 3.16: Bus System Classification

Label Description

Bus Urban Station
BPT Bus Urban Station (with all building type options enabled)

Bus Fuel Facility
BFF Bus Fuel Facility (different combinations for with or without anchored
components and/or with or without backup power)

Bus Dispatch Facility
BDF Bus Dispatch Facility (different combinations for with or without anchored
components and/or with or without backup power)
Bus Maintenance Facility
BMF Bus Maintenance Facilities (with all building type options enabled)

3.3.4.2.Ports and Harbors
Port and harbor transportation systems consist of waterfront structures, cranes/cargo
handling equipment, warehouses and fuel facilities. The inventory data required for
ports and harbors analysis include the geographical location, classification and
replacement cost of the port and harbor system facilities.
3.3.4.3. Classification
Table 3.17 describes the various classes of port and harbor transportation system
components. For more details on how to classify these components refer to section
7.5 of Chapter 7.

337
HazusMHTechnicalManual
Table 3.17: Port and Harbor System Classification
Label Description

Waterfront Structures
PWS Waterfront Structures

Cranes/Cargo Handling Equipment
PEQ1 Stationary Port Handling Equipment
PEQ2 Rail Mounted Port Handling Equipment

Warehouses
PWH Port Warehouses (with all building type options enabled)

Fuel Facility
PFF Port Fuel Facility Facility (different combinations for with or without
anchored components and/or with or without backup power)
3.3.4.4.Ferry
A ferry transportation system consists of waterfront structures, passenger terminals,
fuel facilities, dispatch facilities and maintenance facilities. The inventory data
required for ferry systems analysis include the geographical location, classification
and replacement cost of ferry system facilities.
3.3.4.5. Classification
Table 3.18 describes the various classes of ferry transportation system components.
For more details on how to classify these components refer to section 7.6 of Chapter
7.

Table 3.18: Ferry System Classification

Label Description
Water Front Structures
FWS Ferry Waterfront Structures
Ferry Passenger Terminals
FPT Passenger Terminals (with all building type options enabled)
Ferry Fuel Facility
FFF Ferry Fuel Facility (different combinations for with or without anchored
components and/or with or without backup power)
Ferry Dispatch Facility
FDF Ferry Dispatch Facility (different combinations for with or without
anchored components and/or with or without backup power)

Ferry Maintenance Facility
FMF Piers and Dock Facilities (with all building type options enabled)
3.3.5. Airports

An airport transportation system consists of control towers, runways, terminal
buildings, parking structures, fuel facilities, and maintenance and hangar facilities.
The inventory data required for airports analysis include the geographical location,
classification and replacement cost of airport facilities.
338
Chapter3InventoryData:CollectionandClassification
3.3.5.1. Classification

Table 3.19 describes the various classes of airport system components. For more
details on how to classify these components refer to section 7.7 of Chapter 7.


Table 3.19: Airport System Classification

Label Description

Airport Control Towers
ACT Airport Control Tower (with all building type options enabled)

Airport Terminal Buildings
ATB Airport Terminal Building (with all building type options enabled)
Airport Parking Structures
APS Airport Parking Structure (with all building type options enabled)
Fuel Facilities
AFF Airport Fuel Facility (different combinations for with or without anchored
components and/or with or without backup power)
Airport Maintenance & Hangar Facility
AMF Airport Maintenance & Hangar Facility (with all building type options enabled)
ARW Airport Runway
Airport Facilities - Others
AFO Gliderport, Seaport, Stolport, Ultralight or Baloonport Facilities
AFH Heliport Facilities


3.4. Direct Damage Data - Lifeline Utility Systems

Lifeline utility systems include potable water, waste water, oil, natural gas, electric
power and communication systems. This section describes the classification of
lifeline utility system and their components, and data required to provide damage and
loss estimates.

3.4.1. Potable Water System

A potable water system consists of pipelines, water treatment plants, wells, storage
tanks and pumping stations. The inventory data required for potable water systems
analysis include the geographical location and classification of system components.
The analysis also requires the replacement cost for facilities and the repair cost for
pipelines.
3.4.1.1. Classification

Table 3.20 describes the various classes of potable water system components. For
more details on how to classify these components refer to section 8.1 of Chapter 8.

339
HazusMHTechnicalManual
Table 3.20: Potable Water System Classification

Label Description
Pipelines
PWP1 Brittle Pipe
PWP2 Ductile Pipe
Pumping Plants
PPPL
Large Pumping Plant ( > 50 MGD ) [different combinations for with or
without anchored components]
PPPM
Medium Pumping Plant ( 10 to 50 MGD ) [different combinations for with
or without anchored components]
PPPS
Small Pumping Plant ( < 10 MGD ) [different combinations for with or
without anchored components]
Wells
PWE Wells
Water Storage Tanks (Typically, 0.5 MGD to 2 MGD)
PSTAS Above Ground Steel Tank
PSTBC Buried Concrete Tank
PSTGC On Ground Concrete Tank
PSTGS On Ground Steel Tank
PSTGW On Ground Wood Tank
Water Treatment Plants
PWTL
Large WTP ( > 200 MGD ) [different combinations for with or without
anchored components]
PWTM
Medium WTP ( 50-200 MGD ) [different combinations for with or without
anchored components]
PWTS
Small WTP ( < 50 MGD ) [different combinations for with or without
anchored components]


3.4.2. Waste Water

A waste water system consists of pipelines, waste water treatment plants and lift
stations. The inventory data required for waste water systems analysis include the
geographical location and classification of system components. The analysis also
requires the replacement cost for facilities and the repair cost for pipelines.
3.4.2.1. Classification

Table 3.21 describes the various classes of waste water system components. For
more details on how to classify these components refer to section 8.2 of Chapter 8.

340
Chapter3InventoryData:CollectionandClassification
Table 3.21: Waste Water System Classification

Label Description
Buried Pipelines
WWP1 Brittle Pipe
WWP2 Ductile Pipe
Waste Water Treatment Plants
WWTL
Large WWTP ( > 200 MGD ) [different combinations for with or without
anchored components]
WWTM
Medium WWTP ( 50-200 MGD ) [different combinations for with or
without anchored components]
WWTS
Small WWTP ( < 50 MGD ) [different combinations for with or without
anchored components]
Lift Stations
WLSL
Large Lift Stations ( > 50 MGD ) [different combinations for with or
without anchored components]
WLSM
Medium Lift Stations ( 10 MGD - 50 MGD) [different combinations for
with or without anchored components]
WLSS
Small Lift Stations ( < 10 MGD ) [different combinations for with or
without anchored components]


3.4.3. Oil Systems

An oil system consists of pipelines, refineries, pumping plants and tank farms. The
inventory data required for oil systems analysis include the geographical location and
classification of system components. The analysis also requires the replacement cost
for facilities and the repair cost for pipelines.

3.4.3.1. Classification

Table 3.22 describes the various classes of oil system components. For more details
on how to classify these components refer to section 8.3 of Chapter 8.

341
HazusMHTechnicalManual
Table 3.22: Oil System Classification

Label Description
Pipelines
OIP1 Welded Steel Pipe with Gas Welded Joints
OIP2 Welded Steel Pipe with Arc Welded Joints
Refineries
ORFL
Large Refinery ( > 500,000 lb./day ) [different combinations for with or without
anchored components]
ORFM
Medium Refinery ( 100,000 - 500,000 lb./ day) [different combinations for with
or without anchored components]
ORFS
Small Refinery ( < 100,000 lb./day ) [different combinations for with or without
anchored components]
Pumping Plants
OPP
Pumping Plant [different combinations for with or without anchored
components]
Tank Farms
OTF
Tank Farms with Anchored Tanks [different combinations for with or without
anchored components]

3.4.4. Natural Gas Systems

A natural gas system consists of pipelines and compressor stations. The inventory
data required for natural gas systems analysis include the geographical location and
classification of system components. The analysis also requires the replacement cost
for facilities and the repair cost for pipelines.

3.4.4.1. Classification

Table 3.23 describes the various classes of natural gas system components. For more
details on how to classify these components refer to section 8.4 of Chapter 8.

Table 3.23: Natural Gas System Classification

Label Description

Buried Pipelines
NGP1 Welded Steel Pipe with Gas Welded Joints
NGP2 Welded Steel Pipe with Arc Welded Joints

Compressor Stations
NGC Compressor Stations [different combinations for with or without
anchored components]

342
Chapter3InventoryData:CollectionandClassification
3.4.5. Electric Power

An electric power system consists of substations, distribution circuits, generation
plants and transmission towers. The inventory data required for electric power
systems analysis include the geographical location, classification and replacement
cost of the facilities.
3.4.5.1. Classification

Table 3.24 describes the various classes of electric power system components. For
more details on how to classify these components refer to section 8.5 of Chapter 8.

Table 3.24: Electric Power System Classification

Label Description
Transmission Substations
ESSL
Low Voltage (115 KV) Substation [different combinations for with or
without anchored components]
ESSM
Medium Voltage (230 KV) Substation [different combinations for with or
without anchored components]
ESSH
High Voltage (500 KV) Substation [different combinations for with or
without anchored components]
Distribution Circuits
EDC
Distribution Circuits (either Seismically Designed Components or Standard
Components)
Generation Plants
EPPL
Large Power Plants ( > 500 MW ) [different combinations for with or without
anchored components]
EPPM
Medium Power Plants ( 100 - 500 MW ) [different combinations for with or
without anchored components]
EPPS
Small Power Plants ( < 100 MW ) [different combinations for with or without
anchored components]


3.4.6. Communication

In the loss estimation methodology, a communication system consists of telephone
central offices. The inventory data required for communication systems analysis
include the geographical location and the classification. The analysis also requires
the replacement cost of the facilities.
3.4.6.1. Classification

Table 3.25 describes the various classes of central offices. For more details on how
to classify these components refer to section 8.6 of Chapter 8.

343
HazusMHTechnicalManual
Table 3.25: Communication Classification

Label Description

Central Offices
CCO Central Offices (different combinations for with or without anchored
components and/or with or without backup power)
Stations or Transmitters
CBR AM or FM radio stations or transmitters
CBT TV stations or transmitters
CBW Weather stations or transmitters
CBO Other stations or transmitters


3.5. Hazardous Materials Facilities

Hazardous material facilities contain substances that can pose significant hazards
because of their toxicity, radioactivity, flammability, explosiveness or reactivity.
Significant casualties or property damage could occur form a small number or even a
single hazardous materials release induced by an earthquake, and the consequence of
an earthquake-caused release can vary greatly according to the type and quantity of
substance released, meteorological conditions and timeliness and effectiveness of
emergency response. Similarly to the case of critical faculties with a potential for
high loss, such as large dams, the methodology does not attempt to estimate losses
caused by earthquake which caused hazardous materials releases. Thus, the
hazardous materials module of Hazus is limited to inventory data concerning the
location and nature of hazardous materials located at various sites. Section 11.1.2
describes the scheme used to define the degree of danger of hazardous materials.

3.6. Direct Economic and Social Loss

In this section, information related to inventory data required to determine direct
economic and social loss is presented. The two main databases used to determine
direct economic and social loss are demographic and building square footage
databases.

3.6.1. Demographics Data

The census data are used to estimate direct social loss due to displaced households,
casualties due to earthquakes, and the estimation quality of building space (square
footage) for certain occupancy classes. The Census Bureau collects and publishes
statistics about the people of the United States based on the constitutionally required
census every 10 years, which is taken in the years ending in "0" (e.g., 1990). The
Bureau's population census data describes the characteristics of the population
including age, income, housing and ethnic origin.

344
Chapter3InventoryData:CollectionandClassification
The census data were processed for all of the census tracts in the United States, and
29 fields of direct importance to the methodology were extracted and stored. These
fields are shown in Table 3.26 and are supplied as default information with the
methodology. The population information is aggregated to a census tract level.
Census tracts are divisions of land that are designed to contain 2500-8000 inhabitants
with relatively homogeneous population characteristics, economic status and living
conditions. Census tract divisions and boundaries change only once every ten years.
Census tract boundaries never cross county boundaries, and all the area within a
county is contained within one or more census tracts. This characteristic allows for a
unique division of land from country to state to county to census tract. Each Census
tract is identified by a unique 11 digit number. The first two digits represent the
tract's state, the next three digits represent the tract's county, while the last 6 digits
identify the tract within the county. For example, a census tract numbered
10050505800 would be located in Delaware (10) in Sussex County (050).



345
HazusMHTechnicalManual
Table 3.26: Demographics Data Fields and Usage

Module Usage
Description of Field Shelter Casualty Occupancy
Class
Lifelines
Total Population in Census Tract
* *
Total Household in Census Tract
* *
Total Number of People in General Quarter
*
Total Number of People < 16 years old
*
Total Number of People 16-65 years old
*
Total Number of People > 65 years old
*
Total Number of People - White
*
Total Number of People - Black
*
Total Number of People - Native American
*
Total Number of People - Asian
*
Total Number of People - Hispanic
*
Total # of Households with Income < $10,000
*
Total # of Households with Income $10 - $20K
*
Total # of Households with Income $20 - $30K
*
Total # of Households with Income $30 - $40K
*
Total # of Households with Income $40 - $50K
*
Total # of Households with Income $50 - $60K
*
Total # of Households with Income $60 - $75K
*
Total # of Households with Income $75 - $100K
*
Total # of Households with Income > $100k
*
Total in Residential Property during Day
*
Total in Residential Property at Night
*
Hotel Occupants
*
Vistor Population
*
Total Working Population in Commercial Industry
*
Total Working Population in Industrial Industry
*
Total Commuting at 5 PM
*
Total Number of Students in Grade School
*
Total Number of Students in College/University
*
Total Owner Occupied - Single Household Units
* *
Total Owner Occupied - Multi-Household Units
* *
Total Owner Occupied - Multi-Household Structure
* *
Total Owner Occupied - Mobile Homes
* *
Total Renter Occupied - Single Household Units
* *
Total Renter Occupied - Multi-Household Units
* *
Total Renter Occupied - Multi-Household Structure
* *
Total Renter Occupied - Mobile Homes
* *
Total Vacant - Single Household Units
*
Total Vacant - Multi-Household Units
*
Total Vacant - Multi-Household Structure
*
Total Vacant - Mobile Homes
*
Structure Age <40 years
*
Structure Age >40 years
*

346
Chapter3InventoryData:CollectionandClassification
3.6.2. Default Occupancy Class Square Foot Inventory

The default square footage estimates for occupancy classes RES1, 2,3,5, are based on
census data on the number of dwelling units or the number of people for that
occupancy class. Table 3.27 provides the conversion factors for these occupancy
classes. These conversion factors are obtained from expert opinion and modifications
to ATC-13 values. The conversion factors were also calibrated against tax assessors
data for region-specific counties. The square foot estimates are calculated using the
following expression:

SFI = UD * CF (3-4)
where,
SFI = building square footage for an occupancy class
UD = unit of data for that occupancy class
CF = conversion factor for that occupancy class (Table 3.27)

The building square footage estimates for the remaining occupancy classes were
obtained using a building square footage inventory database purchased from the Dun
and Bradstreet Company (D&B). The square footage information was classified
based on Standard Industrial Code (SIC) and provided at a census tract resolution.
The SIC codes were mapped to NIBS occupancy classes using the mapping scheme
provided in Table 3.27. There is no default information for occupancy class COM10.

3.7. Indirect Economic Data

The indirect economic data refers to the post-earthquake change in the demand and
supply of products, change in employment and change in tax revenues. The user can
specify the levels of potential increase in imports and exports, supply and product
inventories and unemployment rates.


347
HazusMHTechnicalManual
Table 3.27: Mapping of Standard Industrial Codes, Conversion Factors to
Estimate Occupancy Square Footage and Square Footage Per Occupancy Class

Source of Data
Label Occupancy Class Census Dun and Bradstreet
Unit of
Data
Conversion
Factor
SIC Code
Residential
RES1 Single Family Dwelling # of Units variable
RES2 Mobile Home # of Units 1000 sq. t./unit
RES3 Multi Family Dwelling # of Units 1000 sq. t./unit
RES4 Temporary Lodging 70
RES5 Institutional Dormitory # in Group
Quarters
700 sq.
ft./person

RES6 Nursing Home 8051, 8052, 8059
Commercial
COM1 Retail Trade 52, 53, 54, 55, 56, 57, 59
COM2 Wholesale Trade 42, 50, 51
COM3 Personal/Repair Services 72,75,76,83,88
COM4 Prof./Technical Services 40, 41, 44, 45, 46, 47, 49, 61, 62, 63, 64,
65, 67, 73, 78 (except 7832), 81, 87, 89
COM5 Banks 60
COM6 Hospital 8062, 8063, 8069
COM7 Medical Office/Clinic 80 (except 8051, 8052, 8059, 8062, 8063,
8069)
COM8 Entertainment & Rec. 48, 58, 79, (except 7911), 84
COM9 Theaters 7832, 7911
COM10 Parking
Industrial
IND1 Heavy 22, 24, 26, 32, 34, 35 (except 3571, 3572),
37
IND2 Light 23, 25, 27, 30, 31, 36 (except 3671, 3672,
3674), 38, 39
IND3 Food/Drugs/Chemicals 20, 21, 28, 29
IND4 Metals/Minerals
Processing.
10, 12, 13, 14, 33
IND5 High Technology 3571, 3572, 3671, 3672, 3674
IND6 Construction 15, 16, 17
Agriculture
AGR1 Agriculture 01, 02, 07, 08, 09
Religion/Non/Profit
REL1 Church/ N.P. Offices 86
Government
GOV1 General Services 43, 91, 92 (except 9221, 9224), 93, 94, 95,
96, 97
GOV2 Emergency Response 9221, 9224
Education
EDU1 Schools 82 (except 8221, 8222)
EDU2 Colleges/Universities 8221, 8222

348
Chapter3InventoryData:CollectionandClassification
3.8. References

ATC - 13 (1985). Earthquake Damage Evaluation Data for California, Applied
Technology Council, Redwood City, CA.

FEMA, 1993. "Water Control Infrastructure, National Inventory of Dams 1992,"
FEMA 246, Federal Emergency Management Agency and U.S. Army Corps of
Engineers, Washington, D.C., October 1993.

R.S. Means (2002), Means Square Foot Costs

U.S. Bureau of the Census, May 1991. Standard Tape File 1 (STF-1A).

U.S. Bureau of the Census, May 1992. Standard Tape File 3 (STF-3).


349
HazusMHTechnicalManual

APPENDIX 3A
General Building Stock


Table 3A.1: Distribution Percentage of Floor Area for Specific Occupancy
Classes within each General Occupancy Class




General Occupancy Class

Specific Occupancy Class
RES COM IND AGR REL GOV EDU
No. Label Occupancy Class 1 2 3 4 5 6 7
1 RES1 Single Family Dwelling
2 RES2 Mobile Home
3 RES3 Multi Family Dwelling
4 RES4 Temporary Lodging
5 RES5 Institutional Dormitory
6 RES6 Nursing Home
7 COM1 Retail Trade
8 COM2 Wholesale Trade
9 COM3 Personal and Repair Services
10 COM4 Professional/Technical
11 COM5 Banks
12 COM6 Hospital
13 COM7 Medical Office/Clinic
14 COM8 Entertainment & Recreation
15 COM9 Theaters
16 COM10 Parking
17 IND1 Heavy
18 IND2 Light
19 IND3 Food/Drugs/Chemicals
20 IND4 Metals/Minerals Processing
21 IND5 High Technology
22 IND6 Construction
23 AGR1 Agriculture 100
24 REL1 Church 100
25 GOV1 General Services
26 GOV2 Emergency Response
27 EDU1 Schools
28 EDU2 Colleges/Universities
The relative distribution varies by census tract and is computed directly from the specific
occupancy class square footage inventory. For Agriculture (AGR) and Religion (REL) there
is only one specific occupancy class, therefore the distribution is always 100%.
350
Chapter3InventoryData:CollectionandClassification
Table 3A.2: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, Low Rise, Pre-1950, West Coast*
(after ATC-13, 1985)

Specific Model Building Type
No. Occup. 1 2 3 6 9 10 13 16 19 22 25 26 29 31 34 36
Class W1 W2 S1L S2L S3 S4L S5L C1L C2L C3L PC1 PC2L RM1L RM2L URML MH
1 RES1 For State-Specific Res1 Distribution, Refer to Table 3A.17
2 RES2 100
3 RES3 73 1 1 1 6 3 3 1 9 2
4 RES4 34 2 1 2 1 19 16 3 4 18
5 RES5 20 5 1 1 28 18 6 21
6 RES6 45 10 5 10 20 10
7 COM1 22 2 6 3 20 17 1 6 23
8 COM2 8 3 4 2 41 18 1 3 5 2 13
9 COM3 28 1 1 3 18 7 1 8 33
10 COM4 27 2 1 3 19 15 7 26
11 COM5 27 2 1 3 19 15 7 26
12 COM6 8 5 2 11 11 27 2 1 27 6
13 COM7 25 5 2 10 10 15 2 1 20 10
14 COM8 8 12 1 2 3 16 27 4 5 1 21
15 COM9 5 20 7 15 20 3 10 20
16 COM10 8 8 18 43 7 1 6 3 6
17 IND1 3 29 13 2 2 15 14 7 1 4 2 8
18 IND2 4 14 8 22 1 18 16 1 1 2 13
19 IND3 1 18 8 3 3 20 22 2 3 20
20 IND4 2 24 12 7 2 13 16 2 2 6 14
21 IND5 21 5 5 3 35 2 10 2 15 2
22 IND6 32 3 2 10 18 8 7 13 7
23 AGR1 56 3 2 14 2 9 1 13
24 REL1 22 8 2 21 15 5 8 19
25 GOV1 9 8 1 3 4 12 42 4 6 11
26 GOV2 45 2 37 3 13
27 EDU1 11 6 3 3 21 21 4 9 22
28 EDU2 2 5 10 5 15 20 20 5 18
* Refer to Table 3C.1 for states classifications.
351
HazusMHTechnicalManual
Table 3A.3: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, Low Rise, 1950-1970 , West Coast*
(after ATC-13, 1985)

Specific Model Building Type
No. Occup. 1 2 3 6 9 10 13 16 19 22 25 26 29 31 34 36
Class W1 W2 S1L S2L S3 S4L S5L C1L C2L C3L PC1 PC2L RM1L RM2L URML MH
1 RES1 For State-Specific Res1 Distribution, Refer to Table 3A.18
2 RES2 100
3 RES3 72 1 2 2 1 6 2 8 3 3
4 RES4 55 1 2 2 2 3 11 2 18 1 3
5 RES5 39 3 3 1 8 16 6 18 1 5
6 RES6 70 3 1 1 5 20
7 COM1 34 3 1 3 2 4 13 5 10 1 18 2 4
8 COM2 12 4 5 5 3 3 18 22 1 19 4 4
9 COM3 12 3 5 5 2 3 23 4 12 1 22 4 4
10 COM4 34 3 3 1 2 3 17 5 3 23 4 2
11 COM5 34 3 3 1 2 3 17 5 3 23 4 2
12 COM6 32 5 2 4 3 16 6 28 4
13 COM7 46 13 1 3 3 9 20 5
14 COM8 13 17 12 3 3 13 6 30 3
15 COM9 10 10 30 5 10 5 30
16 COM10 5 8 20 34 5 20 6 2
17 IND1 10 25 30 3 7 14 9 2
18 IND2 8 5 14 17 4 10 5 22 3 12
19 IND3 14 16 6 1 5 17 28 1 10 2
20 IND4 18 25 9 11 10 7 15 3 2
21 IND5 4 9 3 2 4 20 35 3 15 4 1
22 IND6 30 1 15 7 4 20 3 20
23 AGR1 51 4 8 12 2 10 11 2
24 REL1 20 4 1 3 3 24 4 37 4
25 GOV1 21 6 3 2 2 26 5 4 2 27 2
26 GOV2 50 13 7 20 10
27 EDU1 25 3 4 5 4 20 4 2 29 4
28 EDU2 5 2 12 5 20 50 6
* Refer to Table 3C.1 for states classifications.
352
Chapter3InventoryData:CollectionandClassification
Table 3A.4: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, Low Rise, Post-1970, West Coast*
(after ATC-13, 1985)

Specific Model Building Type
No. Occup. 1 2 3 6 9 10 13 16 19 22 25 26 29 31 34 36
Class W1 W2 S1L S2L S3 S4L S5L C1L C2L C3L PC1 PC2L RM1L RM2L URML MH
1 RES1 For State-Specific Res1 Distribution, Refer to Table 3A.19
2 RES2 100
3 RES3 73 2 3 6 1 1 9 5
4 RES4 53 3 2 3 4 13 20 2
5 RES5 33 3 3 6 5 24 23 3
6 RES6 70 5 5 20
7 COM1 26 9 1 2 1 6 10 1 15 5 21 3
8 COM2 8 4 1 3 4 2 12 41 3 19 3
9 COM3 13 3 2 2 3 3 13 20 5 34 2
10 COM4 35 3 2 1 3 4 15 8 3 24 2
11 COM5 35 3 2 1 3 4 15 8 3 24 2
12 COM6 31 6 1 1 7 4 13 7 28 2
13 COM7 47 16 5 4 6 2 20
14 COM8 4 23 8 1 3 2 15 4 1 32 7
15 COM9 5 27 20 12 4 27 5
16 COM10 8 8 6 3 49 3 13 7 3
17 IND1 11 19 28 3 2 1 9 11 3 11 1 1
18 IND2 3 13 9 6 3 10 41 3 12
19 IND3 2 15 10 5 3 12 28 7 18
20 IND4 1 26 18 5 4 1 11 1 12 5 15 1
21 IND5 1 12 8 2 3 10 38 7 17 1 1
22 IND6 30 4 6 11 8 16 6 14 5
23 AGR1 40 8 11 8 3 11 1 15 1 2
24 REL1 23 12 3 1 6 26 1 3 22 3
25 GOV1 8 15 4 3 7 2 32 4 16 9
26 GOV2 40 3 7 23 10 7 3 7
27 EDU1 24 9 6 1 5 3 16 3 4 3 21 5
28 EDU2 5 10 10 5 20 5 40 5
* Refer to Table 3C.1 for states classifications.

353
HazusMHTechnicalManual
Table 3A.5: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, Mid Rise, Pre-1950, West Coast*
(after ATC-13, 1985)

Specific Model Building Type
No. Occupancy 4 7 11 14 17 20 23 27 30 32 35
Class S1M S2M S4M S5M C1M C2M C3M PC2M RM1M RM2M URMM
3 RES3 15 4 5 1 19 25 8 23
4 RES4 18 4 12 1 20 20 8 17
5 RES5 16 1 5 40 20 18
6 RES6 20 5 35 20 10 10
7 COM1 8 6 3 21 34 11 1 16
8 COM2 8 27 53 5 7
9 COM3 18 22 42 5 13
10 COM4 25 7 10 2 22 16 9 9
11 COM5 25 7 10 2 22 16 9 9
12 COM6 18 4 6 1 35 19 8 9
13 COM7 20 5 5 30 20 10 10
14 COM8 25 20 40 5 10
15 COM9 30 10 40 10 10
16 COM10 10 5 2 55 18 3 2 5
17 IND1
18 IND2 10 5 75 10
19 IND3 32 3 1 1 14 41 3 5
20 IND4 25 3 1 9 52 10
21 IND5 35 10 30 5 20
22 IND6 20 80
23 AGR1 25 75
24 REL1 10 90
25 GOV1 30 15 5 3 23 10 4 10
26 GOV2
28 EDU2 10 20 60 3 5 2
* Refer to Table 3C.1 for states classifications.

354
Chapter3InventoryData:CollectionandClassification
Table 3A.6: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, Mid Rise, 1950-1970, West Coast*
(after ATC-13, 1985)

Specific Model Building Type
No. Occup. 4 7 11 14 17 20 23 27 30 32 35
Class S1M S2M S4M S5M C1M C2M C3M PC2M RM1M RM2M URMM
3 RES3 10 15 6 4 37 1 21 6
4 RES4 9 24 9 5 34 1 14 4
5 RES5 6 1 11 9 45 18 10
6 RES6 15 10 15 5 25 25 5
7 COM1 7 25 5 3 31 22 7
8 COM2 21 3 2 34 1 34 5
9 COM3 10 3 28 54 5
10 COM4 17 18 9 9 18 2 23 4
11 COM5 17 18 9 9 18 2 23 4
12 COM6 14 10 14 5 23 3 23 8
13 COM7 15 10 15 5 25 25 5
14 COM8 5 28 52 10 5
15 COM9 5 30 50 10 5
16 COM10 5 8 8 7 39 8 18 7
17 IND1 10 20 40 20 10
18 IND2 15 10 50 20 5
19 IND3 11 4 10 30 20 1 15 9
20 IND4 100
21 IND5 10 5 13 32 30 10
22 IND6
23 AGR1
24 REL1 80 10 10
25 GOV1 15 6 15 11 28 2 18 5
26 GOV2 5 10 10 5 60 10
28 EDU2 20 15 5 35 15 10
* Refer to Table 3C.1 for states classifications.

355
HazusMHTechnicalManual
Table 3A.7: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, Mid Rise, Post-1970, West Coast*
(after ATC-13, 1985)

Specific Model Building Type
No. Occupancy 4 7 11 14 17 20 23 27 30 32 35
Class S1M S2M S4M S5M C1M C2M C3M PC2M RM1M RM2M URMM
3 RES3 9 23 8 10 28 7 12 3
4 RES4 16 28 8 11 18 3 13 3
5 RES5 9 10 11 16 34 4 11 5
6 RES6 25 10 15 10 35 5
7 COM1 34 9 3 12 17 5 15 5
8 COM2 20 17 15 10 8 15 15
9 COM3 11 17 3 10 17 12 17 13
10 COM4 37 10 12 9 15 3 9 5
11 COM5 37 10 12 9 15 3 9 5
12 COM6 25 9 15 10 33 1 6 1
13 COM7 25 10 15 10 35 5
14 COM8 10 90
15 COM9 10 90
16 COM10 4 8 3 4 66 8 6 1
17 IND1
18 IND2
19 IND3 62 5 1 23 4 1 3 1
20 IND4 100
21 IND5 18 14 3 34 13 5 10 3
22 IND6
23 AGR1
24 REL1 5 90 5
25 GOV1 25 11 15 22 12 4 9 2
26 GOV2 25 20 35 20
28 EDU2 20 5 10 25 25 10 5
* Refer to Table 3C.1 for states classifications.

356
Chapter3InventoryData:CollectionandClassification
Table 3A.8: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, High Rise, Pre-1950, West Coast*
(after ATC-13, 1985)

Specific Model Building Type
No. Occupancy 5 8 12 15 18 21 24 28 33
Class S1H S2H S4H S5H C1H C2H C3H PC2H RM2H
3
RES3
39 1 2 8 24 23 3
4
RES4
45 3 3 8 20 18 3
5
RES5
15 5 10 30 40
10
COM4
47 10 4 1 21 16 1
11
COM5
47 10 4 1 21 16 1
12
COM6
56 9 1 1 24 8 1
13
COM7

16
COM10

23
AGR1

25
GOV1
53 5 5 3 30 3 1
28
EDU2
5 5 35 40 15
* Refer to Table 3C.1 for states classifications.

Table 3A.9: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, High Rise, 1950-1970, West Coast*
(after ATC-13, 1985)

Specific Model Building Type
No. Occupancy 5 8 12 15 18 21 24 28 33
Class S1H S2H S4H S5H C1H C2H C3H PC2H RM2H
3
RES3
30 21 6 13 24 3 3
4
RES4
48 10 9 12 19 1 1
5
RES5
20 15 25 30 5 5
10
COM4
40 26 18 6 7 1 2
11
COM5
40 26 18 6 7 1 2
12
COM6
35 27 17 4 15 1 1
13
COM7

16
COM10

23
AGR1

25
GOV1
46 13 22 10 8 1
28
EDU2
35 20 20 25
* Refer to Table 3C.1 for states classifications.
357
HazusMHTechnicalManual
Table 3A.10: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, High Rise, Post-1970, West Coast*
(after ATC-13, 1985)

Specific Model Building Type
No. Occupancy 5 8 12 15 18 21 24 28 33
Class S1H S2H S4H S5H C1H C2H C3H PC2H RM2H
3
RES3
44 6 5 18 20 5 2
4
RES4
56 10 6 16 9 2 1
5
RES5
25 18 20 37
10
COM4
56 10 14 14 5 1
11
COM5
54 10 15 15 5 1
12
COM6
45 6 19 13 17
13
COM7

16
COM10

23
AGR1

25
GOV1
52 14 14 14 6
28
EDU2
30 10 10 50
* Refer to Table 3C.1 for states classifications.

358
Chapter3InventoryData:CollectionandClassification
Table 3A.11: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, Low Rise, Mid-West*

Specific Model Building Type
No. Occup. 1 2 3 6 9 10 13 16 19 22 25 26 29 31 34 36
Class W1 W2 S1L S2L S3 S4L S5L C1L C2L C3L PC1 PC2L RM1
L
RM2
L
URML MH
1 RES1 For State-Specific Res1 Distribution, Refer to Table 3A.20
2 RES2 100
3 RES3 75 2 23
4 RES4 50 3 2 45
5 RES5 20 4 13 2 22 4 2 33
6 RES6 90 10
7 COM1 30 2 4 11 6 7 5 5 2 28
8 COM2 10 2 4 11 6 7 2 10 2 14 2 2 28
9 COM3 30 2 4 11 6 7 5 5 2 28
10 COM4 30 2 4 11 6 7 5 5 2 28
11 COM5 30 2 4 11 6 7 5 5 2 28
12 COM6 2 4 2 2 6 21 4 33 6 2 18
13 COM7 30 2 4 11 6 7 5 5 2 28
14 COM8 30 2 4 11 6 7 5 5 2 28
15 COM9 2 6 14 8 10 4 13 2 22 4 15
16 COM10 2 4 11 6 7 6 21 4 33 6
17 IND1 5 10 25 13 17 2 7 2 12 2 5
18 IND2 10 2 4 11 6 7 2 10 2 14 2 3 27
19 IND3 10 2 4 11 6 7 2 10 2 14 2 3 27
20 IND4 5 10 25 13 17 2 7 2 12 2 5
21 IND5 10 2 4 11 6 7 2 10 2 14 2 2 28
22 IND6 30 2 4 11 6 7 5 5 2 28
23 AGR1 10 2 4 11 6 7 2 10 2 14 2 2 28
24 REL1 30 3 5 3 4 5 5 2 2 41
25 GOV1 15 14 21 7 6 4 3 30
26 GOV2 14 7 17 4 12 3 43
27 EDU1 10 5 12 5 7 11 50
28 EDU2 14 6 12 2 8 11 10 37
* Refer to Table 3C.1 for states classifications.


359
HazusMHTechnicalManual
Table 3A.12: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, Mid Rise, Mid-West*

Specific Model Building Type
No. Occupancy 4 7 11 14 17 20 23 27 30 32 35
Class S1M S2M S4M S5M C1M C2M C3M PC2M RM1M RM2M URMM
3 RES3 10 7 3 14 39 7 2 18
4 RES4 10 7 3 14 37 2 7 2 18
5 RES5 25 62 2 11
6 RES6
7 COM1 3 20 16 6 11 27 2 5 2 8
8 COM2 7 3 14 37 2 7 3 27
9 COM3 3 20 16 6 11 27 2 5 2 8
10 COM4 3 20 16 6 11 27 2 5 2 8
11 COM5 3 20 16 6 11 27 2 5 2 8
12 COM6 3 20 16 6 12 30 2 6 5
13 COM7 3 20 16 6 11 27 2 5 2 8
14 COM8 3 20 16 6 11 27 2 5 2 8
15 COM9
16 COM10 2 14 10 4 17 43 2 8
17 IND1
18 IND2 7 3 14 37 2 7 3 27
19 IND3 7 3 14 37 2 7 3 27
20 IND4
21 IND5 7 3 14 37 2 7 3 27
22 IND6
23 AGR1 7 3 14 37 2 7 3 27
24 REL1 3 20 16 6 11 27 2 5 2 8
25 GOV1 20 24 11 9 5 31
26 GOV2
28 EDU2 7 14 9 13 13 44
* Refer to Table 3C.1 for states classifications.

360
Chapter3InventoryData:CollectionandClassification
Table 3A.13: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, High Rise, Mid-West*

Specific Model Building Type
No. Occup. 5 8 12 15 18 21 24 28 33
Class S1H S2H S4H S5H C1H C2H C3H PC2H RM2H
3
RES3
3 13 4 16 44 7 7 6
4
RES4
3 13 4 16 44 7 7 6
5
RES5
26 74
10
COM4
7 29 9 12 32 4 4 3
11
COM5
7 29 9 12 32 4 4 3
12
COM6
7 29 9 13 36 2 2 2
13
COM7
7 29 9 12 32 4 4 3
16
COM10
5 19 6 18 52
23
AGR1
2 6 2 16 44 11 11 8
25
GOV1

28
EDU2

* Refer to Table 3C.1 for states classifications.

361
HazusMHTechnicalManual
Table 3A.14: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, Low Rise, East Coast*

Specific Model Building Type
No. Occup. 1 2 3 6 9 10 13 16 19 22 25 26 29 31 34 36
Class W1 W2 S1L S2L S3 S4L S5L C1L C2L C3L PC1 PC2L RM1L RM2L URML MH
1 RES1 For State-Specific Res1 Distribution, Refer to Table 3A.21
2 RES2 100
3 RES3 62 3 2 2 5 4 22
4 RES4 48 5 4 4 8 4 3 3 3 3 15
5 RES5 7 7 6 6 17 6 3 8 6 5 5 24
6 RES6 22 11 8 8 8 3 2 4 3 5 4 22
7 COM1 14 20 15 5 16 3 2 2 4 2 17
8 COM2 10 21 15 7 16 3 2 2 3 4 17
9 COM3 25 7 5 11 5 3 2 2 6 4 30
10 COM4 26 11 8 4 9 4 2 3 5 4 24
11 COM5 13 13 9 13 10 5 3 2 2 5 3 22
12 COM6 2 22 15 18 10 4 2 5 4 3 2 13
13 COM7 24 10 7 15 8 3 2 3 4 4 20
14 COM8 19 19 13 6 15 3 2 2 3 3 15
15 COM9 5 20 13 12 2 16 7 2 3 3 3 2 12
16 COM10 10 7 8 30 11 6 14 12 2
17 IND1 5 22 15 4 2 17 7 3 3 3 3 3 13
18 IND2 10 15 9 15 11 5 3 2 2 4 5 19
19 IND3 7 25 18 3 19 4 2 2 2 3 2 13
20 IND4 7 26 19 3 20 3 2 2 2 3 13
21 IND5 5 25 17 3 2 20 7 3 3 3 2 10
22 IND6 10 21 14 7 2 16 5 2 2 2 2 3 14
23 AGR1 48 8 6 12 7 2 3 2 12
24 REL1 36 4 4 3 2 2 2 7 6 34
25 GOV1 7 24 16 3 19 5 3 2 1 3 3 13
26 GOV2 8 16 11 4 13 8 3 2 4 3 4 5 19
27 EDU1 13 17 13 13 5 3 2 2 5 5 22
28 EDU2 4 18 13 14 8 3 2 4 3 5 4 22
* Refer to Table 3C.1 for states classifications.


362
Chapter3InventoryData:CollectionandClassification
Table 3A.15: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, Mid Rise, East Coast*

Specific Model Building Type
No. Occupancy 4 7 11 14 17 20 23 27 30 32 35
Class S1M S2M S4M S5M C1M C2M C3M PC2M RM1M RM2M URMM
3 RES3 3 4 6 3 14 13 57
4 RES4 9 12 3 18 9 2 11 7 29
5 RES5 7 10 3 23 11 3 12 5 26
6 RES6
7 COM1 23 29 2 8 5 3 5 5 20
8 COM2 23 30 3 8 4 3 5 5 19
9 COM3 10 13 3 5 4 11 10 44
10 COM4 14 19 2 5 7 4 9 7 33
11 COM5 15 21 2 6 8 5 8 6 29
12 COM6 21 27 2 8 12 6 2 7 2 13
13 COM7 15 20 2 5 7 4 9 6 32
14 COM8 22 30 3 8 5 3 5 5 19
15 COM9
16 COM10 10 13 3 38 17 6 11 2
17 IND1
18 IND2 22 28 2 8 10 5 2 6 3 14
19 IND3 25 32 3 9 6 4 4 3 14
20 IND4
21 IND5 24 32 3 9 9 6 5 2 10
22 IND6
23 AGR1 19 25 2 7 4 2 7 6 28
24 REL1 5 9 2 4 3 12 12 53
25 GOV1 24 30 3 9 7 5 5 3 14
26 GOV2
28 EDU2 17 23 2 6 10 5 2 8 4 23
* Refer to Table 3C.1 for states classifications.

363
HazusMHTechnicalManual
Table 3A.16: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, High Rise, East Coast*

Specific Model Building Type
No. Occup. 5 8 12 15 18 21 24 28 33
Class S1H S2H S4H S5H C1H C2H C3H PC2H RM2H
3
RES3
8 21 8 34 17 2 5 5
4
RES4
8 21 8 34 17 2 5 5
5
RES5
6 16 6 40 20 3 5 4
10
COM4
15 36 15 15 8 2 9
11
COM5
15 36 15 15 8 2 9
12
COM6
14 35 14 17 8 2 2 8
13
COM7
15 38 15 14 8 2 8
16
COM10
5 12 5 43 21 4 6 4
23
AGR1
7 4 18 20 42 9
25
GOV1

28
EDU2

* Refer to Table 3C.1 for states classifications.


Table 3A.17: Distribution Percentage of Floor Area for Model Building Types
within RES1 Building Occupancy Class, Pre-1950, West Coast

Model Building Type
State
State State
1 9 13 19 29 34
FIPS* Abbreviation W1 S3 S5L C2L RM1L URML
02
AK Alaska
99 1
04 AZ
Arizona
60 25 16
06 CA
California
99 1 0
08 CO
Colorado
76 15 9
15 HI
Hawaii
92 1 4 3
16 ID
Idaho
95 3 2
30 MT
Montana
98 1 1
35 NM
New Mexico
74 16 10
32 NV
Nevada
97 2 1
41 OR
Oregon
99 1
49 UT
Utah
82 11 7
53 WA
Washington
98 1 1
56 WY
Wyoming
92 5 3
* State FIPS are two digit unique number representative of each state and US territory.
Refer to Table 3C.1 of Appendix C for a complete list of State FIPS.
364
Chapter3InventoryData:CollectionandClassification
Table 3A.18: Distribution Percentage of Floor Area for Model Building Types
within RES1 Building Occupancy Class, 1950-1970, West Coast

Model Building Type
State
State State
1 9 13 19 29 34
FIPS Abbreviation W1 S3 S5L C2L RM1L URML
02
AK Alaska
99 1
04 AZ
Arizona
60 36 4
06 CA
California
99 1 0
08 CO
Colorado
76 21 3
15 HI
Hawaii
92 1 6 1
16 ID
Idaho
95 4 1
30 MT
Montana
98 2
35 NM
New Mexico
74 23 3
32 NV
Nevada
97 3
41 OR
Oregon
99 1
49 UT
Utah
82 16 2
53 WA
Washington
98 2
56 WY
Wyoming
92 7 1


Table 3A.19: Distribution Percentage of Floor Area for Model Building Types
within RES1 Building Occupancy Class, Post-1970, West Coast

Model Building Type
State
State State
1 9 13 19 29 34
FIPS Abbreviation W1 S3 S5L C2L RM1L URML
02
AK Alaska
99 1
04 AZ
Arizona
60 40
06 CA
California
99 1 0
08 CO
Colorado
76 24
15 HI
Hawaii
92 1 7
16 ID
Idaho
95 5
30 MT
Montana
98 2
35 NM
New Mexico
74 26
32 NV
Nevada
97 3
41 OR
Oregon
99 1
49 UT
Utah
82 18
53 WA
Washington
98 2
56 WY
Wyoming
92 8

365
HazusMHTechnicalManual
Table 3A.20: Distribution Percentage of Floor Area for Model Building Types
within RES1 Building Occupancy Class, Mid-West

Model Building Type
State
State State
1 19 34
FIPS Abbreviation W1 C2L URML
05 AR
Arkansas
87 13
19 IA
Iowa
92 8
17 IL
Illinois
77 1 22
18 IN
Indiana
80 20
20 KS
Kansas
91 9
21 KY
Kentucky
88 12
22 LA
Louisiana
89 11
26 MI
Michigan
86 14
27 MN
Minnesota
95 1 4
29 MO
Missouri
76 24
28 MS
Mississippi
94 6
38 ND
North Dakota
98 2
31 NE
Nebraska
89 1 10
39 OH
Ohio
76 24
40 OK
Oklahoma
71 29
46 SD
South Dakota
97 3
47 TN
Tennessee
90 10
48 TX
Texas
100
55 WI
Wisconsin
90 10


366
Chapter3InventoryData:CollectionandClassification
Table 3A.21: Distribution Percentage of Floor Area for Model Building Types
within RES1 Building Occupancy Class, East Coast

Model Building Type
State
State State
1 19 34
FIPS Abbreviation W1 C2L URML
01 AL
Alabama
95 5
09 CT
Connecticut
96 4
11 DC
District of Columbia
21 3 76
10 DE
Delaware
71 1 28
12 FL
Florida
25 5 70
13 GA
Georgia
93 7
25 MA
Massachusetts
96 4
24 MD
Maryland
71 1 28
23 ME
Maine
99 1
37 NC
North Carolina
90 10
33 NH
New Hampshire
97 1 2
34 NJ
New Jersey
91 9
36 NY
New York
85 1 14
42 PA
Pennsylvania
66 34
44 RI
Rhode Island
98 2
45 SC
South Carolina
92 8
51 VA
Virginia
75 25
50 VT
Vermont
96 2 2
54 WV
West Virginia
72 28



367
HazusMHTechnicalManual






APPENDIX 3B
Essential Facilities


Table 3B.1: Distribution Percentage of Floor Area for Specific Occupancy
Classes within each General Occupancy Class

General Occupancy Class
Specific Occupancy Class Medical Care Emergency Response Schools
No. Label Occupancy Class 1 2 3
1 EFHS Small Hospital X
2 EFHM Medium Hospital X
3 EFHL Large Hospital X
4 EFMC Medical Clinics X
5 EFFS Fire Station X
6 EFPS Police Station X
7
EFEO
Emergency
Operation Centers
X
8 EFS1 Grade Schools X
9
EFS2
Colleges/
Universities
X


Table 3B.2: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, Low Rise, Pre-1950, West Coast*
(after ATC-13, 1985)

Specific Model Building Type
No. Occup. 1 2 3 6 9 10 13 16 19 22 25 26 29 31 34
Class W1 W2 S1L S2L S3 S4L S5L C1L C2L C3L PC1 PC2L RM1L RM2L URML
1 EFHS 8 5 2 11 11 27 2 1 27 6
2 EFHM 8 5 2 11 11 27 2 1 27 6
3 EFHL 8 5 2 11 11 27 2 1 27 6
4 EFMC 8 5 2 11 11 27 2 1 27 6
5 EFFS 45 2 37 3 13
6 EFPS 45 2 37 3 13
7 EFEO 45 2 37 3 13
8 EFS1 11 6 3 3 21 21 4 9 22
9 EFS2 2 5 10 5 15 20 20 5 18


368
Chapter3InventoryData:CollectionandClassification
Table 3B.3: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, Low Rise, 1950-1970, West Coast*
(after ATC-13, 1985)

Specific Model Building Type
No. Occup. 1 2 3 6 9 10 13 16 19 22 25 26 29 31 34
Class W1 W2 S1L S2L S3 S4L S5L C1L C2L C3L PC1 PC2L RM1L RM2L URML
1 EFHS 32 5 2 4 3 16 6 28 4
2 EFHM 32 5 2 4 3 16 6 28 4
3 EFHL 32 5 2 4 3 16 6 28 4
4 EFMC 32 5 2 4 3 16 6 28 4
5 EFFS 50 13 7 20 10
6 EFPS 50 13 7 20 10
7 EFEO 50 13 7 20 10
8 EFS1 25 3 4 5 4 20 4 2 29 4
9 EFS2 5 2 12 5 20 50 6
* Refer to Table 3C.1 for states classifications.


Table 3B.4: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, Low Rise, Post-1970, West Coast*
(after ATC-13, 1985)

Specific Model Building Type
No. Occup. 1 2 3 6 9 10 13 16 19 22 25 26 29 31 34
Class W1 W2 S1L S2L S3 S4L S5L C1L C2L C3L PC1 PC2L RM1L RM2L URML
1 EFHS 31 6 1 1 7 4 13 7 28 2
2 EFHM 31 6 1 1 7 4 13 7 28 2
3 EFHL 31 6 1 1 7 4 13 7 28 2
4 EFMC 31 6 1 1 7 4 13 7 28 2
5 EFFS 40 3 7 23 10 7 3 7
6 EFPS 40 3 7 23 10 7 3 7
7 EFEO 40 3 7 23 10 7 3 7
8 EFS1 24 9 6 1 5 3 16 3 4 3 21 5
9 EFS2 5 10 10 5 20 5 40 5


369
HazusMHTechnicalManual
Table 3B.5: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, Mid Rise, Pre-1950, West Coast*
(after ATC-13, 1985)

Specific Model Building Type
No. Occupancy 4 7 11 14 17 20 23 27 30 32 35
Class S1M S2M S4M S5M C1M C2M C3M PC2M RM1M RM2M URMM
1 EFHS 18 4 6 1 35 19 8 9
2 EFHM 18 4 6 1 35 19 8 9
3 EFHL 18 4 6 1 35 19 8 9
4 EFMC 18 4 6 1 35 19 8 9
5 EFFS
6 EFPS
7 EFEO
9 EFS2 10 20 60 3 5 2
* Refer to Table 3C.1 for states classifications.


Table 3B.6: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, Mid Rise, 1950-1970, West Coast*
(after ATC-13, 1985)

Specific Model Building Type
No. Occupancy 4 7 11 14 17 20 23 27 30 32 35
Class S1M S2M S4M S5M C1M C2M C3M PC2M RM1M RM2M URMM
1 EFHS 14 10 14 5 23 3 23 8
2 EFHM 14 10 14 5 23 3 23 8
3 EFHL 14 10 14 5 23 3 23 8
4 EFMC 14 10 14 5 23 3 23 8
5 EFFS 5 10 10 5 60 10
6 EFPS 5 10 10 5 60 10
7 EFEO 5 10 10 5 60 10
9 EFS2 20 15 5 35 15 10


370
Chapter3InventoryData:CollectionandClassification
Table 3B.7: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, Mid Rise, Post-1970, West Coast*
(after ATC-13, 1985)

Specific Model Building Type
No. Occupancy 4 7 11 14 17 20 23 27 30 32 35
Class S1M S2M S4M S5M C1M C2M C3M PC2M RM1M RM2M URMM
1 EFHS 25 9 15 10 33 1 6 1
2 EFHM 25 9 15 10 33 1 6 1
3 EFHL 25 9 15 10 33 1 6 1
4 EFMC 25 9 15 10 33 1 6 1
5 EFFS 25 20 35 20
6 EFPS 25 20 35 20
7 EFEO 25 20 35 20
9 EFS2 20 5 10 25 25 10 5
* Refer to Table 3C.1 for states classifications.


Table 3B.8: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, High Rise, Pre-1950, West Coast*
(after ATC-13, 1985)

Specific Model Building Type
No. Occupancy 5 8 12 15 18 21 24 28 33
Class S1H S2H S4H S5H C1H C2H C3H PC2H RM2H
1 EFHS 56 9 1 1 24 8 1
2 EFHM 56 9 1 1 24 8 1
3 EFHL 56 9 1 1 24 8 1
4 EFMC 56 9 1 1 24 8 1
9 EFS2 5 5 35 40 15


Table 3B.9: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, High Rise, 1950-1970, West Coast*
(after ATC-13, 1985)

Specific Model Building Type
No. Occupancy 5 8 12 15 18 21 24 28 33
Class S1H S2H S4H S5H C1H C2H C3H PC2H RM2H
1 EFHS 35 27 17 4 15 1 1
2 EFHM 35 27 17 4 15 1 1
3 EFHL 35 27 17 4 15 1 1
4 EFMC 35 27 17 4 15 1 1
9 EFS2 35 20 20 25

371
HazusMHTechnicalManual

Table 3B.10: Distribution Percentage of Floor Area, for Model Building Types
within Each Building Occupancy Class, High Rise, Post-1970, West Coast*
(after ATC-13, 1985)

Specific Model Building Type
No. Occupancy 5 8 12 15 18 21 24 28 33
Class S1H S2H S4H S5H C1H C2H C3H PC2H RM2H
1 EFHS 45 6 19 13 17
2 EFHM 45 6 19 13 17
3 EFHL 45 6 19 13 17
4 EFMC 45 6 19 13 17
9 EFS2 30 10 10 50
* Refer to Table 3C.1 for states classifications.


Table 3B.11: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, Low Rise, Mid-West*

Specific Model Building Type
No. Occup. 1 2 3 6 9 10 13 16 19 22 25 26 29 31 34
Class W1 W2 S1L S2L S3 S4L S5L C1L C2L C3L PC1 PC2L RM1L RM2L URML
1 EFHS 30 2 4 11 6 7 5 5 2 28
2 EFHM 2 4 2 2 6 21 4 33 6 2 18
3 EFHL 2 4 2 2 6 21 4 33 6 2 18
4 EFMC 30 2 4 11 6 7 5 5 2 28
5 EFFS 14 7 17 4 12 3 43
6 EFPS 14 7 17 4 12 3 43
7 EFEO 14 7 17 4 12 3 43
8 EFS1 10 5 12 5 7 11 50
9 EFS2 14 6 12 2 8 11 10 37


Table 3B.12: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, Mid Rise, Mid-West*

Specific Model Building Type
No. Occupancy 4 7 11 14 17 20 23 27 30 32 35
Class S1M S2M S4M S5M C1M C2M C3M PC2M RM1M RM2M URMM
1 EFHS 3 20 16 6 11 27 2 5 2 8
2 EFHM 3 20 16 6 12 30 2 6 5
3 EFHL 3 20 16 6 12 30 2 6 5
4 EFMC 3 20 16 6 11 27 2 5 2 8
5 EFFS
6 EFPS
7 EFEO
9 EFS2 7 14 9 13 13 44
* Refer to Table 3C.1 for states classifications.
372
Chapter3InventoryData:CollectionandClassification


Table 3B.13: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, High Rise, Mid-West*

Specific Model Building Type
No. Occupancy 5 8 12 15 18 21 24 28 33
Class S1H S2H S4H S5H C1H C2H C3H PC2H RM2H
1 EFHS 7 29 9 12 32 4 4 3
2 EFHM 7 29 9 13 36 2 2 2
3 EFHL 7 29 9 13 36 2 2 2
4 EFMC 7 29 9 12 32 4 4 3
7 EFEO
9 EFS2


Table 3B.14: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, Low Rise, East Coast*

Specific Model Building Type
No. Occup. 1 2 3 6 9 10 13 16 19 22 25 26 29 31 34
Class W1 W2 S1L S2L S3 S4L S5L C1L C2L C3L PC1 PC2L RM1L RM2L URML
1 EFHS 24 10 7 15 8 3 2 3 4 4 20
2 EFHM 2 22 15 18 10 4 2 5 4 3 2 13
3 EFHL 2 22 15 18 10 4 2 5 4 3 2 13
4 EFMC 24 10 7 15 8 3 2 3 4 4 20
5 EFFS 8 16 11 4 13 8 3 2 4 3 4 5 19
6 EFPS 8 16 11 4 13 8 3 2 4 3 4 5 19
7 EFEO 8 16 11 4 13 8 3 2 4 3 4 5 19
8 EFS1 13 17 13 13 5 3 2 2 5 5 22
9 EFS2 4 18 13 14 8 3 2 4 3 5 4 22
* Refer to Table 3C.1 for states classifications.


373
HazusMHTechnicalManual
Table 3B.15: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, Mid Rise, East Coast*

Specific Model Building Type
No. Occupancy 4 7 11 14 17 20 23 27 30 32 35
Class S1M S2M S4M S5M C1M C2M C3M PC2M RM1M RM2M URMM
1 EFHS 15 20 2 5 7 4 9 6 32
2 EFHM 21 27 2 8 12 6 2 7 2 13
3 EFHL 21 27 2 8 12 6 2 7 2 13
4 EFMC 15 20 2 5 7 4 9 6 32
5 EFFS
6 EFPS
7 EFEO
9 EFS2 17 23 2 6 10 5 2 8 4 23


Table 3B.16: Distribution Percentage of Floor Area for Model Building Types
within Each Building Occupancy Class, High Rise, East Coast*

Specific Model Building Type
No. Occupancy 5 8 12 15 18 21 24 28 33
Class S1H S2H S4H S5H C1H C2H C3H PC2H RM2H
1 EFHS 15 38 15 14 8 2 8
2 EFHM 14 35 14 17 8 2 2 8
3 EFHL 14 35 14 17 8 2 2 8
4 EFMC 15 38 15 14 8 2 8
7 EFEO
9 EFS2
Refer to Table 3C.1 for states classifications.

374
Chapter3InventoryData:CollectionandClassification

APPENDIX 3C

States Classifications

Table 3C.1: Regional Distribution of States

State Fips State
Abbreviation
State Name Group
02 AK Alaska West
01 AL Alabama East
05 AR Arkansas Mid-West
04 AZ Arizona West
06 CA California West
08 CO Colorado West
09 CT Connecticut East
11 DC District of Columbia East
10 DE Delaware East
12 FL Florida East
13 GA Georgia East
15 HI Hawaii West
19 IA Iowa Mid-West
16 ID Idaho West
17 IL Illinois Mid-West
18 IN Indiana Mid-West
20 KS Kansas Mid-West
21 KY Kentucky Mid-West
22 LA Louisiana Mid-West
25 MA Massachusetts East
24 MD Maryland East
23 ME Maine East
26 MI Michigan Mid-West
27 MN Minnesota Mid-West
29 MO Missouri Mid-West
28 MS Mississippi Mid-West
30 MT Montana West
37 NC North Carolina East
38 ND North Dakota Mid-West
31 NE Nebraska Mid-West
33 NH New Hampshire East
34 NJ New Jersey East
35 NM New Mexico West
32 NV Nevada West
36 NY New York East
39 OH Ohio Mid-West
40 OK Oklahoma Mid-West
41 OR Oregon West
42 PA Pennsylvania East
44 RI Rhode Island East

375
HazusMHTechnicalManual
Table 3C.1(cont.): Regional Distribution of States

State Fips State
Abbreviation
State Name Group
45 SC South Carolina East
46 SD South Dakota Mid-West
47 TN Tennessee Mid-West
48 TX Texas Mid-West
49 UT Utah West
51 VA Virginia East
50 VT Vermont East
53 WA Washington West
55 WI Wisconsin Mid-West
54 WV West Virginia East
56 WY Wyoming West
60 AS American Samoa West
66 GU Guam West
69 MR Northern Mariana Islands West
72 PR Puerto Rico East
78 VI Virgin Islands East

41
HazusMHTechnicalManual
Chapter 4
Potential Earth Science Hazards (PESH)


Potential earth science hazards (PESH) include ground motion, ground failure (i.e.,
liquefaction, landslide and surface fault rupture) and tsunami/seiche. Methods for
developing estimates of ground motion and ground failure are discussed in the following
sections. Tsunami/seiche can be included in the Methodology in the form of user-
supplied inundation maps as discussed in Chapter 9. The Methodology, highlighting the
PESH component, is shown in Flowchart 4.1.

4.1 Ground Motion

4.1.1 Introduction

Ground motion estimates are generated in the form of GIS-based contour maps and
location-specific seismic demands stored in relational databases. Ground motion is
characterized by: (1) spectral response, based on a standard spectrum shape, (2) peak
ground acceleration and (3) peak ground velocity. The spatial distribution of ground
motion can be determined using one of the following methods or sources:

- Deterministic ground motion analysis (Methodology calculation)
- USGS probabilistic ground motion maps (maps supplied with Hazus-MH)
- Other probabilistic or deterministic ground motion maps (user-supplied maps)

Deterministic seismic ground motion demands are calculated for user-specified scenario
earthquakes (Section 4.1.2.1). For a given event magnitude, attenuation relationships
(Section 4.1.2.3) are used to calculate ground shaking demand for rock sites (Site Class
B), which is then amplified by factors (Section 4.1.2.5) based on local soil conditions
when a soil map is supplied by the user. The attenuation relationships provided with the
Methodology include eight (8) attenuations relationships for the Western United States
(WUS) and four (4) for the Central and Eastern United States (CEUS).

In the Methodologys probabilistic analysis procedure, the ground shaking demand is
characterized by spectral contour maps developed by the United States Geological
Survey (USGS) as part of 2002 update of the National Seismic Hazard Maps (Frankel et.
al, 2002). The Methodology includes maps for eight probabilistic hazard levels: ranging
from ground shaking with a 39% probability of being exceeded in 50 years (100 year
return period) to the ground shaking with a 2% probability of being exceeded in 50 years
(2500 year return period). The USGS maps describe ground shaking demand for rock
(Site Class B) sites, which the Methodology amplifies based on local soil conditions.

42
Chapter 4 Potential Earth Science Hazards (PESH)
8. Lifelines-
Utility
Systems
4. Ground Motion 4. Ground Failure
Direct Physical
Damage
6. Essential and
High Potential
Loss Facilities
12. Debris 10. Fire 15. Economic 14. Shelter 9. Inundation 11. HazMat
16. Indirect
Economic
Losses
Potential Earth Science Hazards
Direct Economic/
Social Losses
Induced Physical
Damage
7. Lifelines-
Transportation
Systems
5. General
Building
Stock
13. Casualities


Flowchart 4.1: Ground Motion and Ground Failure Relationship to other Modules
of the Earthquake Loss Estimation Methodology


User-supplied peak ground acceleration (PGA) and spectral acceleration contour maps
may also be used with Hazus-MH (Section 4.1.2.1). In this case, the user must provide
all contour maps in a pre-defined digital format (as specified in the Users Manual). As
stated in Section 4.1.2.1, the Methodology assumes that user-supplied maps include soil
amplification.

4.1.1.1 Form of Ground Motion Estimates / Site Effects
43
HazusMHTechnicalManual

Ground motion estimates are represented by: (1) contour maps and (2) location-specific
values of ground shaking demand. For computational efficiency and improved accuracy,
earthquake losses are generally computed using location-specific estimates of ground
shaking demand. For general building stock the analysis has been simplified so that
ground motion demand is computed at the centroid of a census tract. However, contour
maps are also developed to provide pictorial representations of the variation in ground
motion demand within the study region. When ground motion is based on either USGS
or user-supplied maps, location-specific values of ground shaking demand are
interpolated between PGA, PGV or spectral acceleration contours, respectively.

Elastic response spectra (5% damping) are used by the Methodology to characterize
ground shaking demand. These spectra all have the same standard format defined by a
PGA value (at zero period) and spectral response at a period of 0.3 second (acceleration
domain) and spectral response at a period of 1.0 second (velocity domain). Ground
shaking demand is also defined by peak ground velocity (PGV).

4.1.1.2 Input Requirements and Output Information

For computation of ground shaking demand, the following inputs are required:

- Scenario Basis - The user must select the basis for determining ground shaking
demand from one of three options: (1) a deterministic calculation, (2) probabilistic
maps, supplied with the Methodology, or (3) user-supplied maps. For deterministic
calculation of ground shaking, the user specifies a scenario earthquake magnitude and
location. In some cases, the user may also need to specify certain source attributes
required by the attenuation relationships supplied with the Methodology.

- Attenuation Relationship - For deterministic calculation of ground shaking, the user
selects an appropriate attenuation relationship from those supplied with the
Methodology. Attenuation relationships are based on the geographic location of the
study region (Western United States vs. Central Eastern United States) and on the
type of fault for WUS sources. WUS regions include locations in, or west of, the
Rocky Mountains, Hawaii and Alaska. Figure 4-1 shows the regional separation of
WUS and CEUS locations as defined by USGS in the development of the National
Seismic Hazard Maps. For WUS sources, the attenuation functions predict ground
shaking based on source type, including: (1) strike-slip (SS) faults, (2) reverse-slip
(R) faults, (3) normal (N) faults (4) Interface (IF) events and (5) Interslab (IS) events.
The Methodology provides combinations of attenuation functions for the WUS and
CEUS, respectively, based on the work by the USGS (Frankel et al., 2002, Frankel,
1997, Klein et al., 1998).

44
Chapter 4 Potential Earth Science Hazards (PESH)


Figure 4.1 Boundaries Between WUS and CEUS Locations.

- Soil Map - The user may supply a detailed soil map to account for local site
conditions. This map must identify soil type using a scheme that is based on, or can
be related to, the site class definitions of the 1997 NEHRP Provisions (Section
4.1.2.4), and must be in pre-defined digital format (as specified in the Users
Manual). In the absence of a soil map, Hazus-MH will amplify the ground motion
demand assuming Site Class D soil at all sites. The user can also modify the assumed
Site Class soil type for all sites by modifying the analysis parameters in Hazus-MH
(i.e. change the Site Class from D to A,B,C or E).

4.1.2 Description of Methods

The description of the methods for calculating ground shaking is divided into five
separate areas:
- Basis for ground shaking (Section 4.1.2.1)
- Standard shape of response spectra (Section 4.1.2.2)
- Attenuation of ground shaking (Section 4.1.2.3)
- Distance measurement used with attenuation relationships (Section 4.1.2.4)
- Amplification of ground shaking - local site conditions (Section 4.1.2.5)
WUS CEUS
45
HazusMHTechnicalManual
4.1.2.1 Basis for Ground Shaking

The methodology supports three options as the basis for ground shaking:

- Deterministic calculation of scenario earthquake ground shaking
- Probabilistic seismic hazard maps (USGS)
- User-supplied seismic hazard maps

Deterministic Calculation of Scenario Earthquake Ground Shaking
For deterministic calculation of the scenario event, the user specifies the location (e.g.,
epicenter) and magnitude of the scenario earthquake. The Methodology provides three
options for selection of an appropriate scenario earthquake location. The user can either:
(1) specify an event based on a database of WUS seismic sources (faults), (2) specify an
event based on a database of historical earthquake epicenters, or (3) specify an event
based on an arbitrary choice of the epicenter. These options are described below.

Seismic Source Database (WUS Fault Map)
For the WUS, the Methodology provides a database of seismic sources (fault segments)
developed by the USGS, the California Department of Mines and Geology (CDMG) and
the Nevada Bureau of Mines and Geology (NBMG). The user accesses the database map
(using Hazus-MH) and selects a magnitude and epicenter on one of the identified fault
segments. The database includes information on fault segment type, location, orientation
and geometry (e.g., depth, width and dip angle), as well as on each fault segments
seismic potential (e.g., maximum moment).

The Methodology computes the expected values of surface and subsurface fault rupture
length. Fault rupture length is based on the relationship of Wells and Coppersmith
(1994) given below:

( ) M b + a = L log
10
(4-1)

where: L is the rupture length (km)
M is the moment magnitude of the earthquake

Table 4.1 Regression Coefficients of Fault Rupture Relationship of Wells and
Coppersmith (1994)
Rupture Type Fault Type a b
Surface Strike Slip -3.55 0.74
Reverse -2.86 0.63
All -3.22 0.69
Subsurface Strike Slip -2.57 0.62
Reverse -2.42 0.58
All -2.44 0.59

46
Chapter 4 Potential Earth Science Hazards (PESH)
Fault rupture is assumed to be of equal length on each side of the epicenter, provided the
calculated rupture length is available in both directions along the specified fault segment.
If the epicenter location is less than one-half of the rupture length from an end point of
the fault segment (e.g., the epicenter is located at or near an end of the fault segment),
then fault rupture length is truncated so that rupture does not extend past the end of the
fault segment. If the calculated rupture length exceeds the length of the fault segment,
then the entire fault segment is assumed to rupture between its end points.

Historical Earthquake Database (Epicenter Map)
The Methodology software provides a database of historical earthquakes developed from
three sources (Composite Earthquake Catalog, 2002, Earthquake Data Base, 2002,
Earthquake Seismicity Catalog, 1996) and contains over 8,000 records. The database has
been sorted to remove historical earthquakes with magnitudes less than 5.0. The user
accesses the database via Hazus-MH and selects a historical earthquake epicenter which
includes location, depth and magnitude information.

For the WUS, the attenuation relationships require the user to specify the type, dip angle
and orientation of the fault associated with the selected epicenter. The Methodology
computes the expected values of surface and subsurface fault rupture length using
Equation (4-1). Fault rupture is assumed to be of equal length on each side of the
epicenter. For the CEUS, the attenuation relationships utilize the epicenter location and
depth.

Arbitrary Event
Under this option, the user specifies a scenario event magnitude and arbitrary epicenter
(using Hazus-MH). For the WUS, the user must also supply the type, dip angle and
orientation of the fault associated with the arbitrary epicenter. The Methodology
computes the fault rupture length based on Equation (4-1) and assumes fault rupture to be
of equal length on each side of the epicenter. For the CEUS the user must supply the
depth of the hypocenter.

Probabilistic Seismic Hazard Maps (USGS)
The Methodology includes probabilistic seismic hazard contour maps developed by the
USGS for the 2002 update of the National Seismic Hazard Maps (Frankel et al., 2002).
The USGS maps provide estimates of PGA and spectral acceleration at periods of 0.3
second and 1.0 second, respectively. Ground shaking estimates are available for eight
hazard levels: ranging from the ground shaking with a 39% probability of being exceeded
in 50 years to ground shakeing with a 2% probability of being exceeded in 50 years. In
terms of mean return periods, the hazard levels range from 100 years to 2500 years.

User-Supplied Seismic Hazard Maps
The Methodology allows the user to supply PGA and spectral acceleration contour maps
of ground shaking in a pre-defined digital format (as specified in the Hazus-MH Users
Manual). This option permits the user to develop a scenario event that could not be
described adequately by the available attenuation relationships, or to replicate historical
earthquakes (e.g., 1994 Northridge Earthquake). The maps of PGA and spectral
47
HazusMHTechnicalManual
acceleration (periods of 0.3 and 1.0 second) must be provided. The Methodology
software assumes these ground motion maps include soil amplification, thus no soil map
is required.

If only PGA contour maps are available, the user can develop the other required maps
based on the spectral acceleration response factors given in Table 4.2.


4.1.2.2 Standard Shape of the Response Spectra

The Methodology characterizes ground shaking using a standardized response spectrum
shape, as shown in Figure 4.2. The standardized shape consists of four parts: peak
ground acceleration (PGA), a region of constant spectral acceleration at periods from
zero seconds to T
AV
(seconds), a region of constant spectral velocity at periods from T
AV

to T
VD
(seconds) and a region of constant spectral displacement for periods of T
VD
and
beyond.

In Figure 4.2, spectral acceleration is plotted as a function of spectral displacement
(rather than as a function of period). This is the format of response spectra used for
evaluation of damage to buildings (Chapter 5) and essential facilities (Chapter 6).
Equation (4-2) may be used to convert spectral displacement (inches), to period (seconds)
for a given value of spectral acceleration (units of g), and Equation (4-3) may be used to
convert spectral acceleration (units of g) to spectral displacement (inches) for a given
value of period.


A
D
S
S
= T 32 . 0 (4-2)


2
8 . 9 T S = S
A D
(4-3)

The region of constant spectral acceleration is defined by spectral acceleration at a period
of 0.3 second. The constant spectral velocity region has spectral acceleration
proportional to 1/T and is anchored to the spectral acceleration at a period of 1 second.
The period, T
AV
, is based on the intersection of the region of constant spectral
acceleration and constant spectral velocity (spectral acceleration proportional to 1/T).
The value of T
AV
varies depending on the values of spectral acceleration that define these
two intersecting regions. The constant spectral displacement region has spectral
acceleration proportional to 1/T
2
and is anchored to spectral acceleration at the period,
T
VD
, where constant spectral velocity transitions to constant spectral displacement.

48
Chapter 4 Potential Earth Science Hazards (PESH)
The period, T
VD
, is based on the reciprocal of the corner frequency, f
c
, which is
proportional to stress drop and seismic moment. The corner frequency is estimated in
J oyner and Boore (1988) as a function of moment magnitude (M). Using J oyner and
Boores formulation, the period T
VD
, in seconds, is expressed in terms of the earthquakes
moment magnitude as shown by the following Equation (4-4):

2
5) (M
c VD
10 = 1/f T

= (4-4)

When the moment magnitude of the scenario earthquake is not known (e.g., when using
USGS maps or user-supplied maps), the period T
VD
is assumed to be 10 seconds (i.e.,
moment magnitude is assumed to be M =7.0).


Figure 4.2 Standardized Response Spectrum Shape

Using a standard response spectrum shape simplifies calculation of response needed in
estimating damage and loss. In reality, the shape of the spectrum will vary depending on
whether the earthquake occurs in the WUS or CEUS, whether it is a large or moderate
size event and whether the site is near or far from the earthquake source. However, the
differences between the shape of an actual spectrum and the standard spectrum tend to be
significant only at periods less than 0.3 second and at periods greater than T
VD
, which do
not significantly affect the Methodologys estimation of damage and loss.

The standard response spectrum shape (with adjustment for site amplification) represents
all site/source conditions, except for site/source conditions that have strong amplification
at periods beyond 1 second. Although relatively rare, strong amplification at periods
beyond 1 second can occur. For example, strong amplification at a period of about 2
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
1.2
1.3
1.4
1.5
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Spectral Displacement (inches)
S
p
e
c
t
r
a
l

A
c
c
e
l
e
r
a
t
i
o
n

(
g
'
s
)
Standard Shape - Site Class B
Typical Shape - Site Class B (WUS)
0.3 sec.
1.0 sec.
T
AV
T
VD
S
A
(Velocity Domain) o 1/T
PGA
49
HazusMHTechnicalManual
seconds caused extensive damage and loss to taller buildings in parts of Mexico City
during the 1985 Michoacan earthquake. In this case, the standard response spectrum
shape would tend to overestimate short-period spectral acceleration and to underestimate
long-period (i.e., greater than 1-second) spectral acceleration.

Inferred Ground Shaking Hazard Information
Certain ground shaking hazard information is inferred from other ground shaking hazard
information when complete hazard data is not available. Inferred data includes the
following:

- Peak ground velocity (PGV) is inferred from 1-second spectral acceleration response
- Spectral acceleration response is inferred from the peak ground acceleration (PGA)
- 0.3-second spectral acceleration response is inferred from 0.2-second response

PGV Inferred from 1-Second Spectral Response
Unless supplied by the user (i.e., as user-supplied PGV maps), peak ground velocity
(inches per second) is inferred from 1-second spectral acceleration, S
A1
(units of g), using
Equation (4-5).

65 . 1 /
2
4 . 386
1
|
.
|

\
|
=
A
S PGV
t
(4-5)

The factor of 1.65 in the denominator of Equation (4-5) represents the amplification
assumed to exist between peak spectral response and PGV. This factor is based on the
median spectrum amplification, as given in Table 2 of Newmark and Hall (1982) for a
5%-damped system whose period is within the velocity-domain region of the response
spectrum.

Spectral Acceleration Response Inferred from Peak Ground Acceleration (PGA)

When a user has maps of PGA only, spectral acceleration for the short periods, SA
S
,
maps are developed from PGA, and spectral acceleration for the long period, SA
L
, is
inferred from short period spectral acceleration, SA
S
, based on the factors given in Table
4.2 for WUS and CEUS rock (Site Class B) locations.

The factors given in Table 4.2 are based on the combination attenuation functions for
WUS and CEUS events (Section 4.1.2.3). These factors distinguish between small-
magnitude and large-magnitude events and between sites that are located at different
distances (i.e. CUES: distance to hypocenter and WUS: distance to fault rupture plane).
The ratios of SA
S
/SA
L
and SA
S
/PGA define the standard shape of the response spectrum
for each of the magnitude/distance combinations of Table 4.2.

Table 4.2 requires magnitude and distance information to determine spectrum
amplification factors. This information would likely be available for maps of observed
earthquake PGA, or scenario earthquake PGA, but is not available for probabilistic maps
410
Chapter 4 Potential Earth Science Hazards (PESH)
of PGA, since these maps are aggregated estimates of seismic hazard due to different
event magnitudes and sources.


Table 4.2 Spectral Acceleration Response Factors

Western United States (WUS) Rock (Site Class B)
Distance (km)
SA
S
/PGA given Magnitude, M: SA
S
/SA
L
given Magnitude, M:
5 6 7 7.5 5 6 7 7.5
10 km 1.5 1.8 1.9 1.9 4.5 2.8 1.9 1.6
25 km 1.5 1.8 1.9 1.9 4.8 3.1 2.1 1.8
50 km 1.4 1.8 1.9 1.9 4.5 2.9 2.0 1.7
75 km 1.4 1.8 1.9 1.8 4.3 2.8 1.8 1.6
Central and Eastern United States (CEUS) Rock (Site Class B)
10 km 0.8 1.1 1.4 1.7 7.7 4.2 3.0 2.7
25 km 0.9 1.2 1.4 1.5 6.9 4.0 2.9 2.6
50 km 1.0 1.4 1.6 1.7 5.2 3.8 2.7 2.4
75 km 1.2 1.5 1.7 1.8 9.2 3.5 2.6 2.4


0.3-Second Spectral Acceleration Response Inferred from 0.2-Second Response
Some of the probabilistic maps developed by the USGS (Frankel et al. 2002), estimate
short-period spectral response for a period of 0.2 second. Spectral response at a period of
0.3 second is calculated by dividing 0.2-second response by a factor of 1.1 for WUS
locations and by dividing 0.2-second response by a factor of 1.4 for CEUS locations.

The factors describing the ratio of 0.2-second and 0.3-second response are based on the
default combinations of WUS and CEUS attenuation functions, described in the next
section, and the assumption that large-magnitude events tend to dominate seismic hazard
at most WUS locations and that small-magnitude events tend to dominate seismic hazard
at most CEUS locations.


4.1.2.3 Attenuation of Ground Shaking

Ground shaking is attenuated with distance from the source using relationships provided
with the Methodology. These relationships define ground shaking for rock (Site Class B)
conditions based on earthquake magnitude, M, and other parameters. These relationships
are used to estimate PGA and spectral demand at 0.3 and 1.0 seconds, and with the
standard response spectrum shape (described in Section 4.1.2.2) fully define 5%-damped
demand spectra at a given location.

411
HazusMHTechnicalManual
The Methodology provides eight WUS and five CEUS attenuation functions (Tables 4.3
and 4.4). The attenuation functions provided with the Methodology are consistent with
those utilized by the USGS in the 2002 update of the the National Seismic Hazard Maps
(Frankel et al., 2002). A complete description for each attenuation function can be found
in the referenced citation.

The WUS relationships should be used for study regions located in, or west of, the Rocky
Mountains, Hawaii and Alaska. The CEUS attenuation relationships should be used for
the balance of the continental United States and Puerto Rico.


Table 4.3 Western United States (WUS) Attenuation Functions

Western US Attenuation Functions Mechanism Distance
Abrahamson and Silva (1997) N,SS,R r
rup

Boore, J oyner and Fumal (1997) N,SS,R r
jb

Campbell & Bozorgnia (2003) N,SS,R r
seis

Sadigh,Chang, Egan, Makdisi, and Young (1997) N,SS,R r
rup

Spudich, J oyner, Lindh, Boore, Margaris and Fletcher (1999) N r
jb

Munson and Thurber (1997) N r
jb

Youngs, Chiou, Silva and Humphrey (1997) IF,IS r
rup

Atkinson and Boore (2002) IF,IS r
rup



Table 4.4 Central and Eastern United States (CEUS) Attenuation Functions

Central and Eastern US Attenuation Functions Distance
Atkinson and Boore (1995) r
hypo

Toro, Abrahamson and Schneider (1997) r
jb

Frankel, Mueller, Barnhard, Perkins, Leyendecker, Dickman, Hooper (1996) r
hypo

Campbell (2002) r
rup

Sommerville, Collins, Abrahamson, Braves, and Saikia (2002) r
jb



The source-to-site distance measure used with the attenuation function will be described
in the following section (Section 4.1.2.4).

412
Chapter 4 Potential Earth Science Hazards (PESH)
Combination Attenuation Relationships (WUS and CEUS)

The Methodology also provides a series of combination attenuation relationships (Table
4.5). These relationships are based on the approach used by the USGS in the 2002 update
of the National Seismic Hazard Maps (Frankel, 2002).


Table 4.5 Combination Attenuation Relationships for the WUS and CEUS

Name Participation Attenuation Functions Weight
WUS Shallow Crustal
Event - Extensional
Abrahamson and Silva (1997): Hanging Wall 0.20
Sadigh,Chang, Egan, Makdisi, and Young (1997) 0.20
Boore, J oyner and Fumal (1997) 0.20
Spudich et al. (1999) 0.20
Campbell & Bozorgnia (2003) 0.20
WUS Shallow Crustal
Event Non-Extensional
Abrahamson and Silva (1997): Hanging Wall 0.25
Sadigh,Chang, Egan, Makdisi, and Young (1997) 0.25
Boore, J oyner and Fumal (1997) 0.25
Campbell & Bozorgnia (2003) 0.25
WUS Cascadia
Subduction Event
Youngs, Chiou, Silva and Humphrey (1997) 0.50
Sadigh,Chang, Egan, Makdisi, and Young (1997) 0.50
WUS Deep Event ( > 35
km in depth)
Youngs, Chiou, Silva and Humphrey (1997) 0.50
Atkinson and Boore (2002) - Global 0.25
Atkinson and Boore (2002) - Cascadia 0.25
WUS Hawaiian Event
Munson and Thurber (1997) 0.50
Boore, J oyner and Fumal (1997) 0.50
CEUS Event
Atkinson and Boore (1997) 0.286
Toro, Abrahamson and Schneider (1997) 0.286
Frankel, Mueller, Barnhard, Perkins et al. (1996) 0.286
Campbell (2002) 0.142
CEUS Characteristic
Event (New Madrid and
Charleston)
Atkinson and Boore (1997) 0.250
Toro, Abrahamson and Schneider (1997) 0.250
Frankel, Mueller, Barnhard, Perkins et al. (1996) 0.250
Campbell (2002) 0.125
Sommerville, Collins, Abrahamson et al. (2002) 0.125


413
HazusMHTechnicalManual
Comparison of CEUS and WUS Gound Shaking Estimates

The combination CEUS attenuation function predict significantly stronger ground
shaking than the combination of WUS attenuation functions for the same scenario
earthquake (i.e., same moment magnitude and distance to source). For example, Table
4.6 compares WUS and CEUS rock (Site Class B) response spectra (standard shape) for a
series of magnitude and distance combinations. As shown in Table 4.6, CEUS spectral
demand is about 1.5 to 2.5 times WUS demand for the short periods and between 1.0 to
1.7 times WUS demand for the long period (T =1.0 seconds).

Table 4.6 Comparison of Ground Shaking Estimates for the WUS and CEUS

Western United States (WUS) Rock (Site Class B)
Distance (km)
SA
S
(E)/SA
S
(W) for a given
Magnitude, M:
SA
L
(E)/SA
L
(W) for a given
Magnitude, M:
5 6 7 7.5 5 6 7 7.5
10 km 1.2 1.4 1.8 2.2 0.7 0.9 1.2 1.3
25 km 1.5 1.7 2.1 2.5 1.0 1.4 1.6 1.7
50 km 1.4 1.8 2.2 2.4 1.2 1.4 1.6 1.7
75 km 1.7 1.8 2.4 2.5 0.8 1.5 1.7 1.7


4.1.2.4 Source-to-Site Distance Measures for Attenuation Functions

The source-to-site distance is an integral part of each attenuation relationship and
characterizes the decrease in ground shaking intensity as the distance from the earthquake
source increases. The distance measures used in the Methodology are described in Table
4.7 and illustrated in Figures 4.3 and 4.4. Figure 4.3 illustrates the distance measures
from a vertical fault plane while Figure 4.4 illustrates the same measure for a dipping
fault. In the Methodology, all distances and fault dimensions are in kilometers.

Table 4.7 Source-to-Site Distance Measures

Distance Description
r
epi
Distance from the site to the earthquake epicenter
r
hypo
Distance from the site to the earthquake hypocenter
r
jb
Distance from the site to the vertical projection of the fault rupture plane
r
rup
Distance from the site to the fault rupture plane.
r
seis

Distance from the site to the seismogenic portion of the fault rupture plane.
Campbell (1997) provides estimates for the depth to the seismogenic
portion of the fault rupture plane.
414
Chapter 4 Potential Earth Science Hazards (PESH)


Figure 4.3 Source-to-Site Distances for Vertical Faults



Figure 4.4 Source-to-Site Distances for Dipping Faults.
r
jb
r
hypo
Hypocenter
r
seis
& r
rup
r
rup
r
seis
r
hypo
r
jb
r
jb = 0
Seismogenic depth
Fault Width (W)
Dip angle
r
hypo
Seismogenic depth
r
jb
r
rup
r
seis
r
hypo
Hypocenter
Depth (d)
415
HazusMHTechnicalManual
4.1.2.5 Amplification of Ground Shaking - Local Site Conditions

Amplification of ground shaking to account for local site conditions is based on the site
classes and soil amplification factors proposed for the 1997 NEHRP Provisions (which
are essentially the same as the 1994 NEHRP Provisions, FEMA 222A, 1995). The
NEHRP Provisions define a standardized site geology classification scheme and specify
soil amplification factors for most site classes. The classification scheme of the NEHRP
Provisions is based, in part, on the average shear wave velocity of the upper 30 meters of
the local site geology, as shown in Table 4.8. Users (with geotechnical expertise) are
required to relate the soil classification scheme of soil maps to the classification scheme
shown in Table 4.8.

Table 4.8 Site Classes (from the 1997 NEHRP Provisions)

Site Site Class Description Shear Wave Velocity (m/sec)
Class Minimum Maximum
A HARD ROCK
Eastern United States sites only
1500
B ROCK

760 1500
C VERY DENSE SOIL AND SOFT ROCK
Untrained shear strength u
s
>2000 psf (u
s
>100
kPa) or N >50 blows/ft
360 760
D STIFF SOILS
Stiff soil with undrained shear strength 1000 psf <
u
s
<2000 psf (50 kPa <u
s
<100 kPa) or 15 <N
<50 blows/ft
180 360
E SOFT SOILS
Profile with more than 10 ft (3 m) of soft clay
defined as soil with plasticity index PI >20,
moisture content w >40% and undrained shear
strength u
s
<1000 psf (50 kPa) (N <15 blows/ft)
180
F SOILS REQUIRING SITE SPECIFIC
EVALUATIONS
1. Soils vulnerable to potential failure or collapse
under seismic loading:
e.g. liquefiable soils, quick and highly sensitive
clays, collapsible weakly cemented soils.
2. Peats and/or highly organic clays
(10 ft (3 m) or thicker layer)
3. Very high plasticity clays:
(25 ft (8 m) or thicker layer with plasticity index >75)
4. Very thick soft/medium stiff clays:
(120 ft (36 m) or thicker layer)


Soil amplification factors are provided in Table 4.9 for Site Classes A, B, C, D and E.
No amplification factors are available for Site Class F, which requires special site-
specific geotechnical evaluation and is not used in the Methodology.
416
Chapter 4 Potential Earth Science Hazards (PESH)

Table 4.9 Soil Amplification Factors

Site Class B
Site Class
Spectral Acceleration
A B C D E
Short-Period, S
AS
(g) Short-Period Amplification Factor, F
A

s 0.25 0.8 1.0 1.2 1.6 2.5
0.50 0.8 1.0 1.2 1.4 1.7
0.75 0.8 1.0 1.1 1.2 1.2
1.0 0.8 1.0 1.0 1.1 0.9
> 1.25 0.8 1.0 1.0 1.0 0.8*
1-Second Period, S
A1
(g) 1.0-Second Period Amplification Factor, F
V

s 0.1 0.8 1.0 1.7 2.4 3.5
0.2 0.8 1.0 1.6 2.0 3.2
0.3 0.8 1.0 1.5 1.8 2.8
0.4 0.8 1.0 1.4 1.6 2.4
> 0.5 0.8 1.0 1.3 1.5 2.0*
* Site Class E amplification factors are not provided in the NEHRP Provisions when S
AS

> 1.0 or S
A1
> 0.4. Values shown with an asterisk are based on judgment.

The NEHRP Provisions do not provide soil amplification factors for PGA or PGV. The
Methodology amplifies rock (Site Class B) PGA by the same factor as that specified in
Table 4.10 for short-period (0.3-second) spectral acceleration, as expressed in Equation
(4-15), and amplifies rock (Site Class B) PGV by the same factor as that specified in
Table 4.10 for 1.0-second spectral acceleration, as expressed in Equations (4-16).

Ai i
F PGA PGA = (4-15)

Vi i
F PGV PGV = (4-16)

where: PGA
i
is peak ground acceleration for Site Class i (in units of g)
PGA is peak ground acceleration for Site Class B (in units of g)
F
Ai
is the short-period amplification factor for Site Class i, as
specified in Table 4.10 for spectral acceleration, S
AS

PGV
i
is peak ground acceleration for Site Class i (in units of g)
PGV is peak ground acceleration for Site Class B (in units of g)
F
Vi
is the 1-second period amplification factor for Site Class i,
as specified in Table 4.10 for spectral acceleration, S
A1

417
HazusMHTechnicalManual
Construction of Demand Spectra
Demand spectra including soil amplification effects are constructed at short-periods using
Equation (4-17) and at long-periods using Equation (4-18). The period, T
AV
, which
defines the transition period from constant spectral acceleration to constant spectral
velocity is a function of site class, as given in Equation (4-19). The period, T
VD
, which
defines the transition period from constant spectral velocity to constant spectral
displacement is defined by Equation (4-4), and is not a function of site class.
Ai AS ASi
F S S = (4-17)

Vi A i A
F S S =
1 1
(4-18)

|
|
.
|

\
|
|
|
.
|

\
|
=
Ai
Vi
AS
A
AVi
F
F
S
S
T
1
(4-19)
where: S
ASi
is short-period spectral acceleration for Site Class i (in units of g)
S
AS
is short-period spectral acceleration for Site Class B (in units of g)
F
Ai
is the short-period amplification factor for Site Class i, as specified in
Table 4.10 for spectral acceleration, S
AS

S
A1i
is 1-second period spectral acceleration for Site Class i (in units of g)
S
A1
is 1-second period spectral acceleration for Site Class B (in units of g)
F
Vi
is the 1-second period amplification factor for Site Class i, as specified
in Table 4.10 for spectral acceleration, S
A1

T
AVi
is the transition period between constant spectral acceleration and
constant spectral velocity for Site Class i (sec).

Figure 4.5 illustrates construction of response spectra for Site Class D (stiff soil) and E
(soft soil) from Site Class B (rock) response spectra. These spectra represent response
(of a 5%-damped, linear-elastic single-degree-of-freedom system) located at a WUS site,
20 km from a magnitude M =7.0 earthquake, as predicted by the default combination of
WUS attenuation relationships. Figure 4.5 shows the significance of soil type on site
response (i.e., increase in site response with decrease in shear wave velocity) and the
increase in the value of the transition period, T
AV
, with decrease in shear wave velocity.












Figure 4.5 Example Construction of Site Class B, C and D Spectra - WUS
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 1 2 3 4 5 6 7 8 9 10
Spectral Displacement (inches)
S
p
e
c
t
r
a
l

A
c
c
e
l
e
r
a
t
i
o
n

(
g
'
s
)
Site Class B (Rock)
Site Class D (Stiff Soil)
Site Class E (Soft Soil)
0.3 sec.
1.0 sec.
SA1 x FVE
SA1 x FVD
SAS x FAE
SAS x FAD
SAS
SA1
418
Chapter 4 Potential Earth Science Hazards (PESH)
4.1.3 Guidance for Expert-Generated Ground Motion Estimation
Ground motion estimation is a sophisticated combination of earth science, engineering
and probabilistic methods and should not be attempted by users, including local
geotechnical engineers, who not have the proper expertise.

For users who dont have the expertise to estimate ground motion and who need guidance
on which existing attenuation function to use, the table below first summarizes the 34
choices that currently exist within Hazus. Note that the dependent attenuation functions
are the cocktail-based ones in Hazus.

# Description
Fault
Mechanism
East or
West
Note
1 Abrahamson and Silva (1997) Hanging Wall N W
2 Abrahamson and Silva (1997) Foot Wall N W
3 Abrahamson and Silva (1997) Hanging Wall S W
4 Abrahamson and Silva (1997) Foot Wall S W
5 Abrahamson and Silva (1997) Hanging Wall R W
6 Abrahamson and Silva (1997) Foot Wall R W
7 Atkinson and Boore (1995) E E
8 Toro et al. (1997) E E
9 Sommerville (2002) E E
10 Frankel (1996) E E
11 Campbell (2002) E E
12 Boore, J oyner and Fumal (1997) N W
13 Boore, J oyner and Fumal (1997) S W
14 Boore, J oyner and Fumal (1997) R W
15 Sadigh et al. (1997) N W
16 Sadigh et al. (1997) S W
17 Sadigh et al. (1997) R W
18 Campbell & Bozorgnia (2003) N W
19 Campbell & Bozorgnia (2003) S W
20 Spudich et al. (1999) N W
21 Youngs et al. (1997) F W
22 Youngs et al. (1997) I W
23 Munson and Thurber (1997) N W Hawaii-specific
24 CEUS Event E E DEPENDENT
25 WUS Shallow Crustal Event - Extensional S W DEPENDENT
26 WUS Hawaiian Event S W DEPENDENT
27 WUS Cascadia Subduction Event F W DEPENDENT
28 WUS Deep Event I W DEPENDENT
29 WUS Shallow Crustal Event - Non Extensional R W DEPENDENT
30 Atkinson and Boore (2002) F W
31 Atkinson and Boore (2002) - Global I W
32 Atkinson and Boore (2002) - Cascadia I W
33 WUS Shallow Crustal Event - Extensional N W DEPENDENT
34 CEUS Characteristic Event E E DEPENDENT

When the user creates a study region, Hazus will recognize whether this region is the east
coast (E) or west-coast (W), and automatically filters from the table above the ones
applicable for that region.

419
HazusMHTechnicalManual
The dependent (cocktail-based) attenuations are what would show up as a default in
Hazus. However, the user can choose a more specific attenuation for a variety of reasons
that include the following:

Understanding the effects of different attenuation functions on the results. This is
in particular very important given that ground motion has the maximum impact
possible on the results.

Simulating and setting up upper bound and lower bound estimates due to ground
motion. In this case, the user needs to know which of the attenuation functions
provide the smallest shaking and which of the attenuation functions provide the
largest shaking. The plots below show examples of ground shaking for the east-
coast attenuation functions. In this case and for a moment magnitude of 8, Toros
attenuation function will yield the highest shaking resulting in higher losses and
Atkinson and Boore will yield the lowest shaking resulting in the lowest losses for
this magnitude.

Magnitude 5 Magnitude 6
Magnitude 7 Magnitude 8

Similar comparisons of ground motion should be done for study regions in the
western US. For example, the plot below shows example peak ground
accelerations for strike-slip type of events with magnitude 7. It is worth noting in
this plot that for this example event, the attenuation function that yields the
0.010
0.100
1.000
0 10 20 30 40 50 60 70
Di stance [km]
E
x
a
m
p
l
e

P
G
A

[
g
]
Atkinson and Boore (1995)
Toro et al. (1997)
Sommerville (2002)
Frankel (1996)
Campbell (2002)
0.010
0.100
1.000
10.000
0 20 40 60 80 100
Di stance [km]
E
x
a
m
p
l
e

P
G
A

[
g
]
Atkinson and Boore (1995)
Toro et al. (1997)
Sommerville (2002)
Frankel (1996)
Campbell (2002)
0.010
0.100
1.000
10.000
0 20 40 60 80 100 120
Di stance [km]
E
x
a
m
p
l
e

P
G
A

[
g
]
Atkinson and Boore (1995)
Toro et al. (1997)
Sommerville (2002)
Frankel (1996)
Campbell (2002)
0.010
0.100
1.000
10.000
0 20 40 60 80 100 120
Di stance [km]
E
x
a
m
p
l
e

P
G
A

[
g
]
Atkinson and Boore (1995)
Toro et al. (1997)
Sommerville (2002)
Frankel (1996)
Campbell (2002)
420
Chapter 4 Potential Earth Science Hazards (PESH)
highest shaking within 40 kilometers from the source of the earthquake also yields
the lowest shaking for areas further than 40 kilometers from the source. The
implication here would be when a user wants to choose a particular attenuation
function he/she needs to consider the distance between the source and the
community/study region for which upper and lower bound losses need to be
determined.


For a scenario in Hawaii, the attenuation function that ought to be used is Munson and
Thurber (1997). The plot below shows example PGA values for the different magnitudes
and distances when this attenuation function is used.


0.010
0.100
1.000
0 10 20 30 40 50 60 70 80
Di stance [km]
E
x
a
m
p
l
e

P
G
A

[
g
]
Abrahamson & Silva (1997)
Boore, J oyner & Fumal (1997)
Sadigh et al. (1997)
Campbell & Bozorgnia (2003)
1
1
.
2
9
1
1
.
3
3
1
1
.
6
8
1
2
.
3
5
1
3
.
2
8
1
4
.
4
4
1
5
.
7
6
1
7
.
2
2
1
8
.
7
7
2
0
.
4
1
2
2
.
1
0
2
3
.
8
4
6.5
7
7.5
8
8.5
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1.00
G
r
o
u
n
d

A
c
c
.

(
g
)
Distance from EQ Source (km)
EQ Magnitude
at Source
Munson & Thurber Function Plot
0.900-1.000
0.800-0.900
0.700-0.800
0.600-0.700
0.500-0.600
0.400-0.500
0.300-0.400
0.200-0.300
0.100-0.200
Ground Acc. (g)
INITIAL ZONE
RAMPED ZONE
CAPPED ZONE
421
HazusMHTechnicalManual
Finally, it is important for the user to understand that attenuation functions (e.g., Youngs
et. al. attenuation) in subduction zones must not be used under the arbitrary option near
areas of surface faults.


4.2 Ground Failure
4.2.1 Introduction

Three types of ground failure are considered: liquefaction, landsliding and surface fault
rupture. Each of these types of ground failure are quantified by permanent ground
deformation (PGD). Methods and alternatives for determining PGD due to each mode of
ground failure are discussed below.

4.2.1.1 Scope

The scope of this section is to provide methods for evaluating the ground failure hazards
of: (a) liquefaction, (b) landsliding, and (c) surface fault rupture. The evaluation of the
hazard includes the probability of the hazard occurring and the resulting ground
displacement.

4.2.1.2 Input Requirements and Output Information

Input
Liquefaction
- A geologic map based on the age, depositional environment, and possibly the material
characteristics of the geologic units will be used with Table 4.10 to create a
liquefaction susceptibility map
- Groundwater depth map is supplied with a default depth of 5 feet.
- Earthquake Moment Magnitude (M)
Landsliding
- A geologic map, a topographic map, and a map with ground water conditions will be
used with Table 4.15 to produce a landslide susceptibility map
- Earthquake Moment Magnitude (M)
Surface Fault Rupture
- Location of the surface trace of a segment of an active fault that is postulated
to rupture during the scenario earthquake

Output
Liquefaction, and Landsliding
- Aerial depiction map depicting estimated permanent ground deformations.
Surface Fault Rupture
- No maps are generated, only site-specific demands are determined.

4.2.2 Description of Methods

4.2.2.1 Liquefaction

422
Chapter 4 Potential Earth Science Hazards (PESH)
4.2.2.1.1 Background

Liquefaction is a soil behavior phenomenon in which a saturated soil looses a substantial
amount of strength due to high excess pore-water pressure generated by and accumulated
during strong earthquake ground shaking.

Youd and Perkins (1978) have addressed the liquefaction susceptibility of various types
of soil deposits by assigning a qualitative susceptibility rating based upon general
depositional environment and geologic age of the deposit. The relative susceptibility
ratings of Youd and Perkins (1978) shown in Table 4.10 indicate that recently deposited
relatively unconsolidated soils such as Holocene-age river channel, flood plain, and delta
deposits and uncompacted artificial fills located below the groundwater table have high to
very high liquefaction susceptibility. Sands and silty sands are particularly susceptible to
liquefaction. Silts and gravels also are susceptible to liquefaction, and some sensitive
clays have exhibited liquefaction-type strength losses (Updike, et. al., 1988).

Permanent ground displacements due to lateral spreads or flow slides and differential
settlement are commonly considered significant potential hazards associated with
liquefaction.

4.2.2.1.2 Liquefaction Susceptibility

The initial step of the liquefaction hazard evaluation is to characterize the relative
liquefaction susceptibility of the soil/geologic conditions of a region or subregion.
Susceptibility is characterized utilizing geologic map information and the classification
system presented by Youd and Perkins (1978) as summarized in Table 4.10. Large-scale
(e.g., 1:24,000 or greater) or smaller-scale (e.g., 1:250,000) geologic maps are generally
available for many areas from geologists at regional U.S. Geological Survey offices, state
geological agencies, or local government agencies. The geologic maps typically identify
the age, depositional environment, and material type for a particular mapped geologic
unit. Based on these characteristics, a relative liquefaction susceptibility rating (e.g., very
low to very high) is assigned from Table 4.10 to each soil type. Mapped areas of
geologic materials characterized as rock or rock-like are considered for the analysis to
present no liquefaction hazard.

Table 4.10 Liquefaction Susceptibility of Sedimentary Deposits (from Youd and
Perkins, 1978)

General
Distribution of
Cohesionless

Likelihood that Cohesionless Sediments when
Saturated would be Susceptible to Liquefaction (by
Age of Deposit)
Type of Deposit


Sediments in
Deposits


< 500 yr
Modern

Holocene
< 11 ka

Pleistocene
11 ka - 2 Ma
Pre-
Pleistocene
> 2 Ma
(a) Continental Deposits
River channel Locally variable Very High High Low Very Low
Flood plain Locally variable High Moderate Low Very Low
423
HazusMHTechnicalManual
Alluvial fan and plain Widespread Moderate Low Low Very Low
Marine terraces and plains Widespread --- Low Very Low Very Low
Delta and fan-delta Widespread High Moderate Low Very Low
Lacustrine and playa Variable High Moderate Low Very Low
Colluvium Variable High Moderate Low Very Low
Talus Widespread Low Low Very Low Very Low
Dunes Widespread High Moderate Low Very Low
Loess Variable High High High Unknown
Glacial till Variable Low Low Very Low Very Low
Tuff Rare Low Low Very Low Very Low
Tephra Widespread High High ? ?
Residual soils Rare Low Low Very Low Very Low
Sebka Locally variable High Moderate Low Very Low
(b) Coastal Zone
Delta Widespread Very High High Low Very Low
Esturine Locally variable High Moderate Low Very Low
Beach
High Wave Energy Widespread Moderate Low Very Low Very Low
Low Wave Energy Widespread High Moderate Low Very Low
Lagoonal Locally variable High Moderate Low Very Low
Fore shore Locally variable High Moderate Low Very Low
(c) Artificial
Uncompacted Fill Variable Very High --- --- ---
Compacted Fill Variable Low --- --- ---

Liquefaction susceptibility maps produced for certain regions [e.g., greater San Francisco
region (ABAG, 1980); San Diego (Power, et. al., 1982); Los Angeles (Tinsley, et. al.,
1985); San J ose (Power, et. al., 1991); Seattle (Grant, et. al., 1991); among others] are
also available and may alternatively be utilized in the hazard analysis.

4.2.2.1.3 Probability of Liquefaction

The likelihood of experiencing liquefaction at a specific location is primarily influenced
by the susceptibility of the soil, the amplitude and duration of ground shaking and the
depth of groundwater. The relative susceptibility of soils within a particular geologic unit
is assigned as previously discussed. It is recognized that in reality, natural geologic
deposits as well as man-placed fills encompass a range of liquefaction susceptibilities due
to variations of soil type (i.e., grain size distribution), relative density, etc. Therefore,
portions of a geologic map unit may not be susceptible to liquefaction, and this should be
considered in assessing the probability of liquefaction at any given location within the
unit. In general, we expect non-susceptible portions to be smaller for higher
susceptibilities. This "reality" is incorporated by a probability factor that quantifies the
proportion of a geologic map unit deemed susceptible to liquefaction (i.e., the likelihood
of susceptible conditions existing at any given location within the unit). For the various
susceptibility categories, suggested default values are provided in Table 4.11.

Table 4.11 Proportion of Map Unit Susceptible to Liquefaction
424
Chapter 4 Potential Earth Science Hazards (PESH)

Mapped Relative Susceptibility Proportion of Map Unit
Very High 0.25
High 0.20
Moderate 0.10
Low 0.05
Very Low 0.02
None 0.00


These values reflect judgments developed based on preliminary examination of soil
properties data sets compiled for geologic map units characterized for various regional
liquefaction studies (e.g., Power, et. al., 1982).

As previously stated, the likelihood of liquefaction is significantly influenced by ground
shaking amplitude (i.e., peak horizontal acceleration, PGA), ground shaking duration as
reflected by earthquake magnitude, M, and groundwater depth. Thus, the probability of
liquefaction for a given susceptibility category can be determined by the following
relationship:

P Liquefaction
P Liquefaction PGA a
K K
P
M w
ml SC
SC
=
=

(4-20)
where
P Liquefaction PGA a
SC
= is the conditional liquefaction probability for a
given susceptibility category at a specified level of peak ground
acceleration (See Figure 4.8)
K
M
is the moment magnitude (M) correction factor (Equation 4-21)
K
w
is the ground water correction factor (Equation 4-22)
P
ml
proportion of map unit susceptible to liquefaction (Table 4.11)

Relationships between liquefaction probability and peak horizontal ground acceleration
(PGA) are defined for the given susceptibility categories in Table 4.12 and also
represented graphically in Figure 4.6. These relationships have been defined based on
the state-of-practice empirical procedures, as well as the statistical modeling of the
empirical liquefaction catalog presented by Liao, et. al. (1988) for representative
penetration resistance characteristics of soils within each susceptibility category (See
Section 4.2.3.2.3) as gleaned from regional liquefaction studies cited previously. Note
that the relationships given in Figure 4.6 are simplified representations of the
relationships that would be obtained using Liao, et al. (1988) or empirical procedures.


425
HazusMHTechnicalManual
Peak Horizontal Ground Acceleration, PGA (g)
P
[
L
|
P
G
A
=
a
]
0
0.25
0.5
0.75
1
0 0.1 0.2 0.3 0.4 0.5 0.6
Very High
High
Moderate
Low
Very Low


Figure 4.6 Conditional Liquefaction Probability Relationships for Liquefaction
Susceptibility Categories (after Liao, et. al., 1988).

426
Chapter 4 Potential Earth Science Hazards (PESH)
Table 4.12 Conditional Probability Relationship for Liquefaction
Susceptibility Categories
Susceptibility Category
| |
P Liquefaction PGA a =
Very High 0 s 9.09 a - 0.82 s 1.0
High 0 s 7.67a - 0.92 s 1.0
Moderate 0 s 6.67a -1.0 s 1.0
Low 0 s 5.57a -1.18 s 1.0
Very Low 0 s 4.16a - 1.08 s 1.0
None 0.0

The conditional liquefaction probability relationships presented in Figure 4.6 were
developed for a M =7.5 earthquake and an assumed groundwater depth of five feet
Correction factors to account for other moment magnitudes (M) and groundwater depths
are given by Equations 4-21 and 4-22 respectively. These modification factors are well
recognized and have been explicitly incorporated in state-of-practice empirical
procedures for evaluating the liquefaction potential (Seed and Idriss, 1982; Seed, et. al.,
1985; National Research Council, 1985). These relationships are also presented
graphically in Figures 4.7 and 4.8. The magnitude and groundwater depth corrections are
made automatically in the methodology. The modification factors can be computed using
the following relationships:
K
m
= + 00027 00267 02055 29188
3 2
. . . . M M M (4-21)

K 0.022d
w w
= + 093 . (4-22)

where: K
m
is the correction factor for moment magnitudes other than M=7.5;
K
w
is the correction factor for groundwater depths other than five feet;
M represents the magnitude of the seismic event, and;
d
w
represents the depth to the groundwater in feet.

427
HazusMHTechnicalManual
Earthquake Magnitude, M
K
m




0.0
1.0
2.0
4 5 6 7 8


Figure 4.7 Moment Magnitude (M) Correction Factor for Liquefaction Probability
Relationships (after Seed and Idriss, 1982).

Depth to Groundwater, dw (feet)
K
w





0.0
1.0
2.0
0 10 20 30 40


Figure 4.8 Ground Water Depth Correction Factor for Liquefaction Probability
Relationships.

428
Chapter 4 Potential Earth Science Hazards (PESH)
4.2.2.1.4 Permanent Ground Displacements

Lateral Spreading
The expected permanent ground displacements due to lateral spreading can be
determined using the following relationship:

| | ( )
| |
E PGD K E PGD PGA PL
SC SC
a = =
A
/ (4-23)
where
( )
| |
E PGD PGA PL /
SC
a = is the expected permanent ground displacement for a
given susceptibility category under a specified level of
normalized ground shaking (PGA/PGA(t)) (Figure 4.9)
PGA(t) is the threshold ground acceleration necessary to induce
liquefaction (Table 4.13)
K
A
is the displacement correction factor given by Equation
4-24

This relationship for lateral spreading was developed by combining the Liquefaction
Severity Index (LSI) relationship presented by Youd and Perkins (1987) with the ground
motion attenuation relationship developed by Sadigh, et. al. (1986) as presented in J oyner
and Boore (1988). The ground shaking level in Figure 4.9 has been normalized by the
threshold peak ground acceleration PGA(t) corresponding to zero probability of
liquefaction for each susceptibility category as shown on Figure 4.6. The PGA(t) values
for different susceptibility categories are summarized in Table 4.13.

The displacement term, ( )
| |
E PGD PGA PL /
SC
a = , in Equation 4-23 is based on M

=
7.5 earthquakes. Displacements for other magnitudes are determined by modifying this
displacement term by the displacement correction factor given by Equation 4-24. This
equation is based on work done by Seed & Idriss (1982). The displacement correction
factor, K
A
, is shown graphically in Figure 4.10.

K . . . .
A
= + 00086 00914 04698 09835 M M M
3 2
(4-24)

where M is moment magnitude.

429
HazusMHTechnicalManual
0
20
40
60
80
100
0 1 2 3 4 5
PGA/PGA(t)
D
i
s
p
l
a
c
e
m
e
n
t

(
i
n
c
h
e
s
)


12x - 12 for 1<PGA/PGA(t)<2
18x - 24 for 2<PGA/PGA(t)<3
70x - 180 for 3<PGA/PGA(t)<4

Figure 4.9 Lateral Spreading Displacement Relationship (after Youd and Perkins,
1978; Sadigh, et. al., 1986).


Table 4.13 Threshold Ground Acceleration (PGA(t)) Corresponding to Zero
Probability of Liquefaction
Susceptibility Category PGA(t)
Very High 0.09g
High 0.12g
Moderate 0.15g
Low 0.21g
Very Low 0.26g
None N/A


430
Chapter 4 Potential Earth Science Hazards (PESH)
Earthquake Magnitude, M
D
i
s
p
l
a
c
e
m
e
n
t

C
o
r
r
e
c
t
i
o
n

F
a
c
t
o
r




0
1
2
4 5 6 7 8


Figure 4.10 Displacement Correction Factor, K
A
, for Lateral Spreading
Displacement Relationships (after Seed & Idriss, 1982).


Ground Settlement

Ground settlement associated with liquefaction is assumed to be related to the
susceptibility category assigned to an area. This assumption is consistent with
relationship presented by Tokimatsu and Seed (1987) that indicate strong correlations
between volumetric strain (settlement) and soil relative density (a measure of
susceptibility). Additionally, experience has shown that deposits of higher susceptibility
tend to have increased thicknesses of potentially liquefiable soils. Based on these
considerations, the ground settlement amplitudes are given in Table 4.14 for the portion
of a soil deposit estimated to experience liquefaction at a given ground motion level. The
uncertainty associated with these settlement values is assumed to have a uniform
probability distribution within bounds of one-half to two times the respective value. It is
noted that the relationship presented by Tokimatsu and Seed (1987) demonstrate very
little dependence of settlement on ground motion level given the occurrence of
liquefaction. The expected settlement at a location, therefore, is the product of the
probability of liquefaction (Equation 4-18) for a given ground motion level and the
characteristic settlement amplitude appropriate to the susceptibility category (Table 4.14).

431
HazusMHTechnicalManual
Table 4.14 Ground Settlement Amplitudes for Liquefaction
Susceptibility Categories

Relative Susceptibility Settlement (inches)
Very High 12
High 6
Moderate 2
Low 1
Very Low 0
None 0

4.2.2.2 Landslide

4.2.2.2.1 Background

Earthquake-induced landsliding of a hillside slope occurs when the static plus inertia
forces within the slide mass cause the factor of safety to temporarily drop below 1.0. The
value of the peak ground acceleration within the slide mass required to just cause the
factor of safety to drop to 1.0 is denoted by the critical or yield acceleration a
c
. This
value of acceleration is determined based on pseudo-static slope stability analyses and/or
empirically based on observations of slope behavior during past earthquakes.

Deformations are calculated using the approach originally developed by Newmark
(1965). The sliding mass is assumed to be a rigid block. Downslope deformations occur
during the time periods when the induced peak ground acceleration within the slide mass
a
is
exceeds the critical acceleration a
c
. The accumulation of displacement is illustrated in
Figure 4.11. In general, the smaller the ratio (below 1.0) of a
c
to a
is
, the greater is the
number and duration of times when downslope movement occurs, and thus the greater is
the total amount of downslope movement. The amount of downslope movement also
depends on the duration or number of cycles of ground shaking. Since duration and
number of cycles increase with earthquake magnitude, deformation tends to increase with
increasing magnitude for given values of a
c
and a
is
.

4.2.2.2.2 Landslide Susceptibility

The landslide hazard evaluation requires the characterization of the landslide
susceptibility of the soil/geologic conditions of a region or subregion. Susceptibility is
432
Chapter 4 Potential Earth Science Hazards (PESH)
ky1
ky2
ky3
t2 t1
t
Acceleration
t
Velocity
t
t1
t3
t1 t3
Displacement
Note: Critical accelerations in figure are ky1, ky2 and ky3. In applications, critical
acceleration is usually taken as a single value.

Figure 4.11 Integration of Accelerograms to Determine Downslope Displacements
(Goodman and Seed, 1966).
433
HazusMHTechnicalManual
characterized by the geologic group, slope angle and critical acceleration. The
acceleration required to initiate slope movement is a complex function of slope geology,
steepness, groundwater conditions, type of landsliding and history of previous slope
performance. At the present time, a generally accepted relationship or simplified
methodology for estimating a
c
has not been developed.

The relationship proposed by Wilson and Keefer (1985) is utilized in the methodology.
This relationship is shown in Figure 4.12. Landslide susceptibility is measured on a scale
of I to X, with I being the least susceptible. The site condition is identified using three
geologic groups and groundwater level. The description for each geologic group and its
associated susceptibility is given in Table 4.15. The groundwater condition is divided
into either dry condition (groundwater below level of the sliding) or wet condition
(groundwater level at ground surface). The critical acceleration is then estimated for the
respective geologic and groundwater conditions and the slope angle. To avoid
calculating the occurrence of landsliding for very low or zero slope angles and critical
accelerations, lower bounds for slope angles and critical accelerations are established.
These bounds are shown in Table 4.16. Figure 4.12 shows the Wilson and Keefer
relationships within these bounds.

Slope Angle (degrees)
A
c

-

C
r
i
t
i
c
a
l

A
c
c
e
l
e
r
a
t
i
o
n

(
g
)






0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0 5 10 15 20 25 30 35 40 45 50 55
C (Wet)
C (Dry)
B (Wet)
A (Wet)
B (Dry)
A (Dry)

Figure 4.12 Critical Acceleration as a Function of Geologic Group and Slope Angle
(Wilson and Keefer, 1985).

434
Chapter 4 Potential Earth Science Hazards (PESH)
Table 4.15 Landslide Susceptibility of Geologic Groups

Geologic Group Slope Angle, degrees
0-10 10-15 15-20 20-30 30-40 >40
(a) DRY (groundwater below level of sliding)
A
Strongly Cemented Rocks (crystalline
rocks and well-cemented sandstone,
c
'
=300 psf, |
'
= 35
o
)
None None I II IV VI
B
Weakly Cemented Rocks and Soils (sandy
soils and poorly cemented sandstone,
c
'
=0, |
'
= 35
o
)
None III IV V VI VII
C
Argillaceous Rocks (shales, clayey soil,
existing landslides, poorly compacted fills,
c
'
=0 |
'
= 20
o
)
V VI VII IX IX IX
(b) WET (groundwater level at ground surface)
A
Strongly Cemented Rocks (crystalline
rocks and well-cemented sandstone, c
'

=300 psf, |
'
= 35
o
)
None III VI VII VIII VIII
B
Weakly Cemented Rocks and Soils (sandy
soils and poorly cemented sandstone, c
'
=0,
|
'
= 35
o
)
V VIII IX IX IX X
C
Argillaceous Rocks (shales, clayey soil,
existing landslides, poorly compacted fills,
c
'
=0 |
'
= 20
o
)
VII IX X X X X


Table 4.16 Lower Bounds for Slope Angles and Critical Accelerations for
Landsliding Susceptibility

Slope Angle, degrees Critical Acceleration (g)
Group Dry Conditions Wet Conditions Dry Conditions Wet Conditions
A 15 10 0.20 0.15
B 10 5 0.15 0.10
C 5 3 0.10 0.05

As pointed out by Wieczorek et al. (1985), the relationships in Figure 4.12 are
conservative representing the most landslide-susceptible geologic types likely to be found
in the geologic group. Thus, in using this relationship further consideration must be
given to evaluating the probability of slope failure as discussed in Section 4.2.2.2.3.

In Table 4.17, landslide susceptibility categories are defined as a function of critical
acceleration. Then, using Wilson and Keefer's relationship in Figure 4.14 and the lower
bound values in Table 4.16, the susceptibility categories are assigned as a function of
geologic group, groundwater conditions, and slope angle in Table 4.15. Tables 4.15 and
4.17 thus define the landslide susceptibility.

435
HazusMHTechnicalManual
Table 4.17 Critical Accelerations (a
c
) for Susceptibility Categories

Susceptibility
Category
None I II III IV V VI VII VIII IX X
Critical
Accelerations (g)
None 0.60 0.50 0.40 0.35 0.30 0.25 0.20 0.15 0.10 0.05

4.2.2.2.3 Probability of Having a Landslide-Susceptible Deposit

Because of the conservative nature of the Wilson and Keefer (1985) correlation, an
assessment is made of the percentage of a landslide susceptibility category that is
expected to be susceptible to landslide. Based on Wieczorek et al. (1985), this
percentage is selected from Table 4.18 as a function of the susceptibility categories.
Thus, at any given location, there is a specified probability of having a landslide-
susceptible deposit, and landsliding either occurs or does not occur within susceptible
deposits depending on whether the induced peak ground acceleration a
is
exceeds the
critical acceleration a
c
.

Table 4.18 Percentage of Map Area Having a Landslide-Susceptible Deposit

Susceptibility
Category
None I II III IV V VI VII VIII IX X
Map Area 0.00 0.01 0.02 0.03 0.05 0.08 0.10 0.15 0.20 0.25 0.30


4.2.2.2.4 Permanent Ground Displacements

The permanent ground displacements are determined using the following expression:

| |
| | E PGD E d a a n
is is
= / (4-25)

where
| | E d a
is
/ is the expected displacement factor (Figure 4.14)
a
is
is the induced acceleration (in decimal fraction of g's)
n is the number of cycles (Equation 4-26).

A relationship between number of cycles and earthquake moment magnitude (M) based
on Seed and Idriss (1982) is shown in Figure 4.13 and can be expressed as follows.

n . . . . = + 03419 55214 336154 707692
3 2
M M M (4-26)

The induced peak ground acceleration within the slide mass, a
is
, represents the average
peak acceleration within the entire slide mass. For relatively shallow and laterally small
slides, a
is
is not significantly different than the induced peak ground surface acceleration
436
Chapter 4 Potential Earth Science Hazards (PESH)
a
i
. For deep and large slide masses a
is
is less than a
i
. For many applications a
is
may be
assumed equal to the accelerations predicted by the peak ground acceleration attenuation
relationships being used for the loss estimation study. Considering also that topographic
amplification of ground motion may also occur on hillside slopes (which is not explicitly
incorporated in the attenuation relationships), the assumption of a
is
equal to a
i
may be
prudent. The user may specify a ratio a
is
/a
i
less than 1.0. The default value is 1.0.

Magnitude
N
u
m
b
e
r

o
f

C
y
c
l
e
s




0
5
10
15
20
25
30
5 5.5 6 6.5 7 7.5 8 8.5


Figure 4.13 Relationship between Earthquake Moment Magnitude
and Number of Cycles.

A relationship derived from the results of Makdisi and Seed (1978) is used to calculate
downslope displacements. In this relationship, shown in Figure 4.14, the displacement
factor d/a
is
is calculated as a function of the ratio a
c
/a
is
. For the relationship shown in
Figure 4.14, the range in estimated displacement factor is shown and it is assumed that
there is a uniform probability distribution of displacement factors between the upper and
lower bounds.

437
HazusMHTechnicalManual
ac/ais
D
i
s
p
l
a
c
e
m
e
n
t

F
a
c
t
o
r
,

d
/
a
i
s

(
c
m
/
c
y
c
l
e
)


0.01
0.1
1
10
100
0 0.2 0.4 0.6 0.8 1
Upper Bound
Lower Bound

Figure 4.14 Relationship between Displacement Factor and Ratio of
Critical Acceleration and Induced Acceleration.


4.2.2.3 Surface Fault Rupture

4.2.2.3.1 Permanent Ground Displacements

The correlation between surface fault displacement and earthquake moment magnitude
(M) developed by Wells and Coppersmith (1994) is used. The maximum displacement is
given by the relationship shown in Figure 4.16. It is assumed that the maximum
displacement can potentially occur at any location along the fault, although at the ends of
the fault, displacements must drop to zero. The relationship developed by Wells and
Coppersmith based on their empirical data set for all types of faulting (strike slip, reverse
and normal) is used. It is considered that this relationship provides reasonable estimates
for any type of faulting for general loss estimation purposes. The uncertainty in the
maximum displacement estimate is incorporated in the loss estimation analysis. The log
of the standard deviation of estimate is equal to 0.35 which is equivalent to a factor of
about 2 in the displacement estimate at the plus-or-minus one standard deviation level.

The median maximum displacement (MD) is given by the following relationship:

log( ) . . ( ) MD = + 526 079 M (4-27)

where M is moment magnitude.


438
Chapter 4 Potential Earth Science Hazards (PESH)
Moment Magnitude
M
a
x

D
i
s
p
l
a
c
e
m
e
n
t

i
n

m



0.01
0.1
1
10
100
5 5.5 6 6.5 7 7.5 8
Median
+Std. Dev.
-Std. Dev.


Figure 4.15 Relationship for Estimating Maximum Surface Fault Displacement.

It has been observed that displacements along a fault vary considerably in amplitude from
zero to the maximum value. Wells and Coppersmith found that the average displacement
along the fault rupture segment was approximately equal to one-half the maximum
displacement. This is equivalent to a uniform probability distribution for values of
displacement ranging from zero to the maximum displacement. As a conservative
estimate, a uniform probability distribution from one-half of the maximum fault
displacement to the maximum fault displacement is incorporated in the loss estimation
methodology for any location along the fault rupture.


4.2.3 Guidance for Expert-Generated Ground Failure Estimation

This section provides guidance for users who wish to use more refined methods and data
to prepare improved estimates of ground failure. It is assumed that such users would be
geotechnical experts with sufficient expertise in ground failure prediction to develop site-
specific estimates of PGD based on regional/local data.

4.2.3.1 Input Requirements and Output Information

4.2.3.1.1 Liquefaction

Input
- A map delineating areas of equal susceptibility (i.e., similar age, deposition, material
properties, and ground water depth)
- Probability distribution of susceptibility variation within each area
439
HazusMHTechnicalManual
- Relationships between liquefaction probability and ground acceleration for each
susceptible area
- Maps delineating topographic conditions (i.e., slope gradients and/or free-face
locations) and susceptible unit thicknesses
- Relationships between ground displacements (i.e., lateral spreading and settlement),
and ground acceleration for each susceptible unit, including probability distribution
for displacement; they may vary within a given susceptible unit depending on
topographic and liquefied zone thickness conditions

Output
- Contour maps depicting liquefaction hazard and associated potential ground
displacements

4.2.3.1.2 Landsliding

Input

- A map depicting areas of equal critical or yield acceleration a
c
(i.e., the values of
peak ground acceleration within the slide mass required to just initiate landsliding,
that is, reduce the factor of safety to 1.0 at the instant of time a
c
occurs)
- The probability distribution for a
c
within each area
- The ratio between induced peak ground surface acceleration, a
i
, and the peak ground
acceleration within the slide mass a
is
(note: could be a constant ratio or could vary
for different areas). The value a
is
/a
i
s 1. The default ratio is 1.0
- Relationships between landslide displacement d induced acceleration a
ic
and initial or
yield acceleration a
c
including the probability distribution for d. Different
relationships can be specified for different areas. The default relationship between
the displacement factor d/a
is
and a
c
/a
is
is shown in Figure 4.14.

Output
- Contour maps depicting landsliding hazard and permanent ground displacements

4.2.3.1.3 Surface Fault Rupture

Input
- Predictive relationship for the maximum amount of fault displacement
- Specification of regions of the fault having lower maximum displacements
- Specifying other than the default relationship for the probability distribution between
minimum and maximum amounts of fault rupture displacement

Output
- Amount of fault displacement at locations along the fault trace

440
Chapter 4 Potential Earth Science Hazards (PESH)
4.2.3.2 Liquefaction

4.2.3.2.1 Background

The key for the user in defining analysis inputs is understanding the interrelationship
among factors that significantly influence occurrence of liquefaction and associated
ground displacement phenomena.

During earthquake ground shaking, induced cyclic shear creates a tendency in most soils
to change volume by rearrangement of the soil-particle structure. In loose soils, this
volume change tendency is to compact or densify the soil structure. For soils such as fine
sands, silts and clays, permeability is sufficiently low such that undrained conditions
prevail and no or insignificant volume change can occur during the ground shaking. To
accommodate the volume decrease tendency, the soil responds by increases of pore-water
pressure and corresponding decreases of intergranular effective stress. The relationship
between volume change tendency and pore-water increase is described by Martin, et al.
(1975). Egan and Sangrey (1978) discuss the relationship among compressibility
characteristics, the potential amount of pore-water pressure generation and the
subsequent loss of strength in various soil materials. In general, more compressible soils
such as plastic silts or clays do not generate excess pore-water pressure as quickly or to as
large an extent as less compressible soils such as sands. Therefore, silty and clayey soils
tend to be less susceptible than sandy soils to liquefaction-type behaviors. Even within
sandy soils, the presence of finer-grained materials affects susceptibility as is reflected in
the correlations illustrated in Figure 4.16 prepared by Seed, et al. (1985) for use in
simplified empirical procedures for evaluating liquefaction potential.

Excess pore-water pressure generation and strength loss potential are also highly
dependent on the density of the soil, as may also be inferred from Figure 4.16. Density
characteristics of soils in a deposit, notably sandy and silty soils, are reflected in
penetration resistance measured, for example, during drilling and sampling an
exploratory boring. Using penetration resistance data to help assess liquefaction hazard
due to an earthquake is considered a reasonable engineering approach (Seed and Idriss,
1982; Seed, et. al., 1985; National Research Council, 1985), because many of the factors
affecting penetration resistance affect the liquefaction resistance of sandy and silty soils
in a similar way and because state-of-practice liquefaction evaluation procedures are
based on actual performance of soil deposits during worldwide historical earthquakes
(e.g., Figure 4.16).


Fi

These
pore
proce
topog
soil.
lique
withi
Later
are ty
igure 4.16 R
e displaceme
water press
ess. Later
graphic featu
Earthquak
fiable mater
in the slope
ral spreading
ypically acco
Relationship
(N
1
)
6
ent hazards a
sure and sig
al spreads
ures (i.e., fre
ke ground-sh
rials by caus
and by shak
g movement
ompanied by
p between C
0
values (M
are direct pr
gnificant str
are ground
ee-faces) and
haking affec
sing seismic
king-induced
ts may be on
y surface fis

Cyclic Stress
M=7.5) (Seed
roducts of th
rength reduc
d failure p
d on gently s
cts the stab
inertia force
d strength re
n the order o
sures and slu
H
s Ratio caus
et al., 1985
he soil behav
ction) produ
henomena
sloping grou
bility of slo
es to be add
eductions in
of inches to
umping. Flo
HazusMHTec
ing Liquefa
).
vior phenome
uced by the
that occur
und underlain
oping groun
ded to gravita
the liquefiab
several feet
ow slides ge
44
chnicalManu

action and
ena (i.e., hig
e liquefactio
near abrup
n by liquefie
nd containin
ational force
ble material
t or more an
enerally occu
41
al
gh
on
pt
ed
ng
es
s.
nd
ur
442
Chapter 4 Potential Earth Science Hazards (PESH)
in liquefied materials found on steeper slopes and may involve ground movements of
hundreds of feet. As a result, flowslides can be the most catastrophic of the liquefaction-
related ground-failure phenomena. Fortunately, flow slides are much less common
occurrences than lateral spreads.

Settlement is a result of the dissipation of excess pore pressure generated by the
rearrangement of loosely compacted saturated soils into a denser configuration during
shaking. Such dissipation will produce volume decreases (termed consolidation or
compaction) within the soil that are manifested at the ground surface as settlement.
Volume changes may occur in both liquefied and non-liquefied zones with significantly
larger contributions to settlement expected to result from liquefied soil. Densification
may also occur in loose unsaturated materials above the ground water table. Spatial
variations in material characteristics may cause such settlements to occur differentially.
Differential ground settlement may also occur near sand boil manifestations due to
liquefied materials being removed from the depths of liquefaction and brought to the
ground surface.

These factors have been discussed briefly in preceding sections and incorporated to the
extent possible in characterizing relationships of Section 4.2.2.1. The challenge to the
user is to translate regional/local data, experience and judgment into defining site-specific
relationships. The following paragraphs offer additional comments regarding various
aspects of that process.

4.2.3.2.2 Susceptibility

Fundamental soil characteristics and physical processes that affect liquefaction
susceptibility have been identified through case histories and laboratory studies.
Depositional environments of sediments and their geologic ages control these
characteristics and processes, as discussed by Youd and Perkins (1978).

The depositional environments of sediments control grain size distribution and, in part,
the relative density and structural arrangement of grains. Grain size characteristics of a
soil influence its susceptibility to liquefaction. Fine sands tend to be more susceptible
than silts and gravels. All cohesionless soils, however, may be considered potentially
liquefiable as the influence of particle size distribution is not thoroughly understood. In
general, cohesive soils that contain more than about 20 percent clay may be considered
nonliquefiable (Seed and Idriss, 1982, present criteria for classifying a soil as
nonliquefiable).

Relative density and structural arrangement of grains (soil structure) greatly influence
liquefaction susceptibility of a cohesionless soil. Soils that have higher relative densities
and more stable soil structure have a lower susceptibility to liquefaction. These factors
may be related to both depositional environment and age. Sediments undisturbed after
deposition (e.g., lagoon or bay deposits) tend to have lower densities and less stable
structures than sediments subjected to wave or current action. With increasing age of a
deposit, relative density may increase as particles gradually work closer together. The
443
HazusMHTechnicalManual
soil structure also may become more stable with age through slight particle reorientation
or cementation. Also, the thickness of overburden sediments may increase with age, and
the increased pressures associated with a thicker overburden will tend to increase the
density of the soil deposit.

An increase in the ratio of effective lateral earth pressure to effective vertical or
overburden earth pressure in a soil has been shown to reduce its liquefaction
susceptibility. Such an increase will occur when overburden is removed by erosion.

In general, it is thought that the soil characteristics and processes that result in a lower
liquefaction susceptibility also result in higher penetration resistance when a soil sampler
is driven into a soil deposit. Therefore, blow count values, which measure penetration
resistance of a soil sampler in a boring, are a useful indicator of liquefaction
susceptibility. Similarly, the resistance from pushing a cone penetrometer into the soil is
a useful indicator of liquefaction susceptibility. An understanding of the depositional
environments and ages of soil units together with penetration resistance data enables
assessment of liquefaction susceptibility.

Additional information helpful to enhancing/refining the susceptibility characterization is
observation of liquefaction and related phenomena during historical earthquakes, as well
as evidence of paleoliquefaction. Although such information does not exist for all
locations and its absence does not preclude liquefaction susceptibility, it is available for
numerous locations throughout the country; for example, in Northern California (Youd
and Hoose, 1978; Tinsley, et. al., 1994). Incorporation of such historical information has
been shown to significantly enhance liquefaction-related loss estimation.

4.2.3.2.3 Liquefaction Probability

As described previously, simplified procedures for evaluating liquefaction potential
presented by Seed, et. al. (1985), as well as probabilistic approach presented by Liao, et
al. (1988), are useful tools for helping to characterize the relationships among
liquefaction probability, peak ground acceleration, duration of shaking (magnitude), and
groundwater depth, etc. A parameter commonly utilized in these procedures is
penetration resistance, which was previously discussed relative to susceptibility. Within
a given geologic unit, experience indicates that subsurface investigations may obtain a
certain scatter in penetration resistance without necessarily any observable trend for
variation horizontally or vertically within that unit. In such cases, a single representative
penetration resistance value is often selected for evaluating the liquefaction potential at
the site. The representative value is very much site-specific and depends on the particular
distribution of penetration resistance values measured. For example, if most of the values
are very close to each other, with a few much higher or lower values, the representative
value might be selected as the value that is close to the mean of the predominant
population of values that are close to each other. On the other hand, if the penetration
resistance values appear to be widely scattered over a fairly broad range of values, a
value near the 33rd percentile might be more appropriate to select (H. B. Seed, personal
444
Chapter 4 Potential Earth Science Hazards (PESH)
communication, 1984). A typical distribution of penetration resistance (N
1
) for a
Holocene alluvial fan deposit (i.e., moderate susceptibility) is shown in Figure 4.17.

The user may elect to eliminate the probabilistic factor that quantifies the proportion of a
geologic map unit deemed susceptible to liquefaction (i.e., the likelihood of susceptible
conditions existing at any given location within the unit) if regional geotechnical data
enables microzonation of susceptibility areas, or define this factor as a probabilistic
distribution, or incorporate the susceptibility uncertainty in defining other liquefaction
probability relationships.


Normalized Blow Count, N1 (blows/foot)
P
e
r
c
e
n
t
a
g
e

o
f

D
a
t
a

P
o
i
n
t
s




E
q
u
a
l

t
o

o
r

L
e
s
s

t
h
a
n

N
1



0
10
20
30
40
50
60
70
80
90
100
0 10 20 30 40 50 60 70 80 90
Note: Based on 178 data points


Figure 4.17 Typical Cumulative Distribution Curve of Penetration Resistance for
Holocene Alluvial Fan Deposits (after Power, et. al., 1992).


4.2.3.2.4 Permanent Ground Displacement

Lateral Spreading

Various relationships for estimating lateral spreading displacement have been proposed,
including the previously utilized Liquefaction Severity Index (LSI) by Youd and Perkins
(1978), and a relationship by Bartlet and Youd (1992), in which they characterize
displacement potential as a function of global earthquake and local site characteristics
(e.g., slope, liquefaction thickness, and grain size distribution). Relationships that are
more site-specific may be developed based on simple stability and deformation analysis
for lateral spreading conditions using undrained residual strengths for liquefied sand
(Seed and Harder, 1990) along with Newmark-type (1965) and Makdisi and Seed (1978)
displacement approaches. To reasonably represent the lateral spreading hazard by either
445
HazusMHTechnicalManual
published relationships or area-specific analyses, generalized information regarding
stratigraphic conditions (i.e., depth to and thickness of the liquefied zone) and
topographic conditions (i.e., ground slope and free-face situations) are required.

Ground Settlement

Relationships for assessing ground settlement are available (e.g., Tokimatsu and Seed,
1978) and are suggested to the user for guidance. In addition, test results presented by
Lee and Albaisa (1974) suggest that the magnitude of volumetric strain following
liquefaction may be dependent on grain-size distribution. Area-specific information
required for developing settlement relationships is similar to that for lateral spreading.

4.2.3.3 Landsliding

4.2.3.3.1 Background

The key assessment is the generation of a map denoting areas of equal landslide
susceptibility and their corresponding values of critical acceleration. This should be
accomplished considering the geographical distribution of facilities at risk in the region
and the types of landsliding that could affect the facilities.

4.2.3.3.2 Landslide Susceptibility

Keefer (1984) and Wilson and Keefer (1985) have identified many different types of
landsliding, ranging from rock falls to deep-seated coherent soil or rock slumps to soil
lateral spreads and flows. For loss estimation purposes, the potential for lateral spreads
and flows should be part of the liquefaction potential assessment rather than the landslide
potential. The significance of other forms of downslope movement depends on the
potential for such movements to damage facilities. The emphasis on characterizing
landslide susceptibility should be on failure modes and locations that pose a significant
risk to facilities. For example, if the potential for rock falls were high (because of steep
terrain and weak rock) but could occur only in undeveloped areas, then it would not be
important to characterize the critical acceleration for this mode of failure. As another
example, in evaluating the probability of landsliding and the amount of displacements as
part of a regional damage assessment for a utility district (Power et al., 1994), it was
assessed that two types of landsliding posed the major risk to the facilities and piping:
activation of existing deep-seated landslide deposits that had been mapped in hillside
areas and that had the potential for disrupting areas in which water lines were located
(landslides often covering many square blocks); and local slumping of roadway sidehill
fills in which water lines were embedded.

Having identified the modes and geographic areas of potential landsliding of significance,
critical acceleration can be evaluated for these modes and areas. It is not necessarily
required to estimate a
c
as a function of slope angle. In some cases, it may be satisfactory
to estimate a
c
and corresponding ranges of values for generalized types of landslides and
subregions, for example, reactivation of existing landslides within a certain subregion or
446
Chapter 4 Potential Earth Science Hazards (PESH)
within the total region. However, it is usually necessary to distinguish between dry and
wet conditions because a
c
is usually strongly dependent on groundwater conditions.

In general, there are two approaches to estimating a
c
: an empirical approach utilizing
observations of landsliding in past earthquakes and corresponding records or estimates of
ground acceleration; and an analytical approach, in which values of a
c
are calculated by
pseudo-static slope stability analysis methods. Often, both approaches may be utilized
(Power, et. al., 1994). When using the analytical approach, the sensitivity of results to
soil strength parameters must be recognized. In assessing strength parameter values and
ranges, it is often useful to back-estimate values, which are operable during static
conditions. Thus, for certain types of geology, slope angles, static performance
observations during dry and wet seasons, and estimates of static factors of safety, it may
be possible to infer reasonable ranges of strength parameters from static slope stability
analyses. For earthquake loading conditions, an assessment should also be made as to
whether the short-term dynamic, cyclic strength would differ from the static strength. If
the soil or rock is not susceptible to strength degradation due to cyclic load applications
or large deformations, then it may be appropriate to assign strength values higher than
static values due to rate of loading effects. On the other hand, values even lower than
static values may be appropriate if significant reduction in strength is expected (such as
due to large-deformation-induced remolding of soil).

4.2.3.3.3 Probability of Landsliding

The probability of landsliding at any location is determined by comparing the induced
peak ground acceleration (adjusted to the value of the peak acceleration in the landslide
mass a
is
) with the assessed distribution for critical acceleration a
c
(Figure 4.19).

4.2.3.3.4 Permanent Ground Displacements

In assessing soil deformations using relationships such as shown in Figure 4.14, it should
be kept in mind that the relationships are applicable to slope masses that exhibit
essentially constant critical accelerations. For cases where significant reduction in
strength may occur during the slope deformation process, these relationships may
significantly underestimate deformations if the peak strength values are used. For
example, deformations cannot be adequately estimated using these simplified correlations
in cases of sudden, brittle failure, such as rock falls or soil or rock avalanches on steep
slopes.

4.2.3.4 Surface Fault Rupture

4.2.3.4.1 Permanent Ground Displacements

Refinements or alternatives that an expert may wish to consider in assessing
displacements associated with surface fault rupture include: a predictive relationship for
maximum fault displacement different from the default relationship (Figure 4.15),
specification of regions of the fault rupture (near the ends) where the maximum fault
447
HazusMHTechnicalManual
displacement is constrained to lower values, and specification of other than the default
relationship for the probability distribution of fault rupture between minimum and
maximum values.


Acceleration
ac
1
a
is
ac
2
P
r
o
b
a
b
i
l
i
t
y

D
e
n
s
i
t
y

f
o
r

a




c


Figure 4.18 Evaluation of Probability of Landsliding.

448
Chapter 4 Potential Earth Science Hazards (PESH)
4.3 References

Abrahamson, N.A. and W.J . Silva, 1997. "Empirical Response Spectal Attenuation
Relationships for Shallow Crustal Earthquakes", Seismological Research Letters, v. 68,
no. 1, pp. 94-127.

Association of Bay Area Governments (ABAG), 1980. "Liquefaction Potential and
Liquefaction Susceptibility Maps", San Francisco Bay region, original scale 1:250,000.

Atkinson G.M. and D.M. Boore, 2002. Empirical Ground-motion Relationships for
Subduction Zone Earthquake and their Application to Cascadia and Other Regions,
Bulletin of the Seismological Society of America, in review.

Atkinson G.M. and D.M. Boore., 1995. Ground-Motion Relations for Eastern North
America, Bulletin of the Seismological Society of America, vol. 85, No. 1, February, pp.
17-30.

Bartlett, S.F., and T.L. Youd, 1992. "Empirical Prediction of Lateral Spread
Displacement", in Hamada, M., and O'Rourke, T. D. (eds.), Proceedings from the Fourth
Japan-U.S. Workshop on Earthquake Resistant Design of Lifeline Facilities and
Countermeasures for Soil Liquefaction, Technical Report NCEER-92-0019, v. I, pp. 351-
365.

Boore, D. M., J oyner, W.B., and T.E. Fumal, 1997. "Equations for Estimating Horizontal
Response Spectra and Peak Acceleration from Western North American Earthquake: a
Summary of Recent Work", Seismological Research Letters, v. 68, pp. 128-153.

Borcherdt, R. D. 1994. "New Developments in Estimating Site Response Effects on
Ground Motion", Proceeding of Seminar on New Developments in Earthquake Ground
Motion Estimation and Implications for Engineering Design Practice ATC-35, Applied
Technology Council, Redwood City, California.

Campbell, K.W., 2002. Prediction of Strong Ground using the Hybrid Empricial
Method: Example Application to Eastern North America, Bulletin of the Seismological
Society of America, in review.

Campbell, K.W., 1997. Empirical Near-Source Attenuation Relationships for
Horizontal and Vertical Components of Peak Ground Acceleration, Peak Ground
Velocity and Pueudo-absolute Acceleration Response Spectra, Seismological Research
Letters, v. 68, no. 1, pp. 154-179.

Campbell, K.W., and Y. Bozorgnia, 2003. Updated Near-Source Ground-Motion
(Attenuation) Relations for the Horizontal and Vertical Components of Peak Ground
Acceleration and Acceleration Response Spectra, Bulletin of the Seismological Society
of America v. 93, no. 1, pp. 314-331, February.

449
HazusMHTechnicalManual
Composite Earthquake Catalog, 2002. Advanced National Seismic System, Northern
California Earthquake Data Center, Berkeley, California.

Earthquake Data Base, 2002. National Earthquake Information Center, United States
Geological Survey, Golden, Colorado.

Earthquake Seismicity Catalog, Volume 1, 1996. National Geophysical Data Center,
National Oceanic & Atmospheric Administration, Boulder, Colorado.

Egan, J .A., and D.A. Sangrey, 1978. "Critical State Model for Cyclic Load Pore
Pressure", Proceedings, Earthquake Engineering and Soil Dynamics Conference and
Exhibit, Pasadena, California, J une 19-21.

Federal Emergency Management Agency, 1992. "FEMA 178 - NEHRP Handbook for
the Seismic Evaluation of Existing Buildings", Washington, D. C., Developed by the
Building Seismic Safety Council (BSSC) for the Federal Emergency Management
Agency (FEMA).

Federal Emergency Management Agency, 1995. FEMA 222A and 223A - NEHRP
Recommended Provisions for Seismic Regulations for New Buildings, 1994 Edition,
Washington, D. C., Developed by the Building Seismic Safety Council (BSSC) for the
Federal Emergency Management Agency (FEMA).

Federal Emergency Management Agency, 1997. 1997 NEHRP Recommended Provisions
for Seismic Regulations for New Buildings, (in press) Washington, D. C., Developed by
the Building Seismic Safety Council (BSSC) for the Federal Emergency Management
Agency (FEMA).

Frankel, A.D., Peterson, M.D., Mueller, C.S., Haller, K.M., Wheleler, R.L., Leyendecker,
E. V., Wesson, R.L., Harmsen, S.C., Cramer, C.H., Perkins, D.M. and K.S. Rukstales,
2002. Documentation for the 2002 Update of the National Seismic Hazard Maps, USGS
Open-File Report 02-420, 110pp.

Frankel, A.D., 1997. Issues and Assumptions for Draft Alaska Seismic Hazard Maps,
United States Geological Survey, Golden, CO.

Goodman, R.E., and H.B. Seed, 1966. "Earthquake-Induced Displacements in Sand
Embankments", Journal of the Soil Mechanics and Foundations Division, Proceedings of
the American Society of Civil Engineers, vol. 92, no. SM2, pp. 125-146.

Grant, W.P., Perkins, W.J . and T.L. Youd, 1991. Evaluation of Liquefaction Potential in
Seattle, Washington - USGS Professional Paper 91-0441T: United States Geological
Survey.

J oyner, W.B., and D.M. Boore, 1988. "Measurement, Characterization, and Prediction of
Strong Ground Motion", Proceedings of Earthquake Engineering & Soil Dynamics II, pp.
450
Chapter 4 Potential Earth Science Hazards (PESH)
43- 102. Park City, Utah, 27 J une 1988. New York: Geotechnical Division of the
American Society of Civil Engineers.

Keefer, D. K.,1984. "Landslides Caused by Earthquakes", Geological Society of America
Bulletin, vol. 95, pp. 406-421.

Klein, F., Frankel, A.D., Mueller, C.S., Wesson, R.L., and P. Okubo, 1998.
Documentation for Draft Seimic-Hazard Maps for the State of Hawaii, United States
Geological Survey.

Lee, K. L., and Albaisa, A., 1974. "Earthquake Induced Settlement in Saturated Sands",
Journal of the Soil Mechanics and Foundation Division, ASCE 100(GTA), pp. 387-406.

Liao, S.S., Veneziano, D., and R.V. Whitman, 1988. "Regression Models for Evaluating
Liquefaction Probability", Journal of Geotechnical Engineering, vol. 114, No. 4, April.

Makdisi, F. I. and H.B. Seed, 1978. "Simplified Procedure for Estimating Dam and
Embankment Earthquake-Induced Deformations", Journal of the Geotechnical
Engineering Division, American Society of Civil Engineers, vol. 104, No. GT7, J uly, pp.
849-867.

Martin G.R., Finn, W.D., and H.B. Seed, 1975. "Fundamentals of Liquefaction Under
Cyclic Loading", Journal of the Geotechnical Engineering Division, ASCE 104(GT5),
pp. 423-438.

Munson, C.G., and C.H. Thurber, 1997. "Analysis of the Attenuation of Strong Ground
Motion on the Island of Hawaii", Bulletin of the Seismological Society of America, vol.
87, No. 4, August, pp. 945-960.

National Research Council, 1985, "Liquefaction of Soils During Earthquakes",
Committee on Earthquake Engineering, Commission on Engineering and Technical
Systems, National Academy Press, Washington, DC.

Newmark, N.M., 1965. "Effects of Earthquakes on Dams and Embankments",
Geotechnique, vol. 15, no. 2, pp. 139-160.

Newmark, N.M. and W.J . Hall, 1982. Earthquake Spectra and Design, EERI
Monograph.

Power, M. S., A.W. Dawson, D.W. Streiff, R.G. Perman, and S. C. Haley, 1982.
Evaluation of Liquefaction Susceptibility in the San Diego, California Urban Area.
Proceedings 3rd International Conference on Microzonation, II, pp. 957-968.

Power, M. S., R. G. Perman, J . R. Wesling, R. R. Youngs, M. K. Shimamoto, 1991.
Assessment of Liquefaction Potential in the San J ose, California Urban Area.
451
HazusMHTechnicalManual
Proceedings 4th International Conference on Microzonation, II, pp. 677-684, Stanford,
CA.

Power, M. S., Taylor, C. L., and Perman, R. C., 1994. "Evaluation of Regional
Landsliding Hazard Potential for Utility District - Eastern San Francisco Bay Area",
Proceedings of the U.S. Fifth National Conference on Earthquake Engineering, Chicago,
Illinois, J uly .

Sadigh, K., Chang C. Y., Egan, J ., Makdisi, F. and R. Youngs, 1997. Attenuation
Relationships for Shallow Crustal Earthquakes based on California Strong Motion Data,
Seismological Research Letters, v. 68, no. 1, pp. 180-189.

Sadigh, K., Egan, J . A., and Youngs, R. R., 1986. "Specification of Ground Motion for
Seismic Design of Long Period Structures", Earthquake Notes, vol. 57, no. 1, p. 13,
relationships are tabulated in J oyner and Boore (1988) and Youngs and others (1987).

Seed, R. B., and Harder, L. F., 1990. "SPT-Based Analysis of Cyclic Pore Pressure
Generation and Undrained Residual Strength", Proceedings of the H. B. Seed Memorial
Symposium, vol. 2, pp. 351-376.

Seed, H. B., and Idriss, I. M., 1971. "Simplified Procedure for Evaluating Soil
Liquefaction Potential", Journal of the Soil Mechanics and Foundations Division, ASCE,
vol. 97, no. SM9, Proceedings Paper 8371, September, pp. 1249-1273.

Seed, H. B., and Idriss, I. M. 1982. "Ground Motions and Soil Liquefaction During
Earthquakes", Earthquake Engineering Research Institute, Oakland, California,
Monograph Series, p. 13.

Seed, H. B., Tokimatsu, K., Harder, L. F., and Chung, R. M., 1985. "Influence of SPT
Procedures in Soil Liquefaction Resistance Evaluations, Journal of Geotechnical
Engineering, American Society of Civil Engineers, , vol. 111, no. 12, p. 1425-1445.

Somerville, P., Collins, N., Abrahamson, R., and C. Saikia, 2001. Ground-motion
Attenuation Relationships for the Central and Eastern United States, final report to the
U.S. Geological Survey.

Spudich, P., J oyner, W.B., Lindh, A.G., Boore, D.M., Margaris, B.M., and J .B. Flecher,
1999. SEA99: A Revised Ground Motion Prediction Relation for use in Extensional
Tectonic Regimes, Bulletin of the Seismological Society of America, vol. 89, pp. 1156-
1170.

Tinsley, J . C., T. L. Youd, D. M. Perkins, and A. T. F. Chen, 1985. Evaluating
Liquefaction Potential, Evaluating Earthquake Hazards in the Los Angeles Region -
USGS Professional Paper 1360. pp. 263-316.

452
Chapter 4 Potential Earth Science Hazards (PESH)
Tokimatsu, A. M., and Seed, H. B., 1987. "Evaluation of Settlements in Sands Due to
Earthquake Shaking", Journal of the Geotechnical Division, American Society of Civil
Engineers, vol. 113, no. 8, pp. 681-878.

Toro, G. R., Abrahamson, N. A. and Schneider, J . F., 1997. Engineering Model of
Strong Ground Motions from Earthquakes in the Central and Eastern United States,
Seismological Research Letters, v. 68, pp.41-57.

Toro, G. R. and McGuire, R. K., 1991. Evaluations of Seismic Hazard at the Paducah
Gaseous Diffusion Plant, Third DOE Natural Phenomena Hazards Mitigation
Conference, pp. 157-165.

Updike, R. G., Egan, J . A., Moriwaki, Y., Idriss, I. M., and Moses, T. L., 1988, "A Model
for Earthquake-Induced Translatory Landslides in Quaternary Sediments", Geological
Society of America Bulletin, vol. 100, May, pp. 783-792.

Wells, D. L. and Coppersmith, K. J ., 1994. "New Empirical Relationships Among
Magnitude, Rupture Length, Rupture Width, and Surface Displacement", Bulletin of the
Seismological Society of America, v 84, pp. 974-1002.

Wieczorek, G. F., Wilson, R. C. and Harp, E. L., 1985. "Map of Slope Stability During
Earthquakes in San Mateo County, California", U.S. Geological Survey Miscellaneous
Investigations Map I-1257-E, scale 1:62,500.

Wilson, R. C., and Keefer D. K., 1985. "Predicting Areal Limits of Earthquake Induced
Landsliding, Evaluating Earthquake Hazards in the Los Angeles Region", U.S.
Geological Survey Professional Paper, Ziony, J . I., Editor, pp. 317-493.

Youd, T. L., and Hoose, S. N., 1978. "Historic Ground Failures in Northern California
Triggered by Earthquakes", U.S. Geological Survey Professional Paper 993, 177 p.

Youd, T. L., and Perkins, D. M., 1978. "Mapping of Liquefaction Induced Ground
Failure Potential", Journal of the Geotechnical Engineering Division, American Society
of Civil Engineers, vol. 104, no. 4, pp. 433-446.

Youd, T. L., and Perkins, D. M., 1987. "Mapping of Liquefaction Severity Index",
Journal of the Geotechnical Engineering Division, American Society of Civil Engineers,
vol. 118, no. 11, pp. 1374-1392.

Youngs R. R., Chiou, S. J ., Silva W. L., and Humphrey J . R., 1997. "Strong Ground
Motion Attenuation Relationships for Subduction Zone Earthquakes", Seismological
Research Letters, J anuary/February.

151
HazusMHTechnicalManual
Chapter 5
Direct Physical Damage - General Building Stock

5.1 Introduction

This chapter describes methods for determining the probability of Slight, Moderate, Extensive
and Complete damage to general building stock. General building stock represents typical
buildings of a given model building type designed to either High-Code, Moderate-Code, or Low-
Code seismic standards, or not seismically designed (referred to as Pre-Code buildings). Chapter
6 describes methods for estimating earthquake damage to essential facilities that include Special
buildings designed and constructed to standards above normal Code provisions. The flowchart of
the overall methodology, highlighting the building damage component and showing its
relationship to other components, is shown in Flowchart 5-1.

5.1.1 Scope

The scope of this chapter includes development of methods for estimation of earthquake damage
to buildings given knowledge of the model building type and an estimate of the level of ground
shaking (or degree of ground failure). Model building types are defined in Section 5.2. The
extent and severity of damage to structural and nonstructural components of a building is
described by one of five damage states: None, Slight, Moderate, Extensive, and Complete.
Damage states are defined in Section 5.3 for each model building type by physical descriptions of
damage to building elements.

This chapter focuses on the development of functions for estimating building damage due to
ground shaking. These building damage functions include: (1) fragility curves that describe the
probability of reaching or exceeding different states of damage given peak building response, and
(2) building capacity (push-over) curves that are used (with damping-modified demand spectra)
to determine peak building response. For use in lifeline damage evaluation, a separate set of
building fragility curves expresses the probability of structural damage in terms of peak ground
acceleration (PGA). Building damage functions for ground shaking are described in Section 5.4
for each model building type.

While ground shaking typically dominates damage to buildings, ground failure can also be a
significant contributor to building damage. Ground failure is characterized by permanent ground
deformation (PGD) and fragility curves are used to describe the probability of reaching different
states of damage given PGD. These fragility curves are similar to, but less detailed than, those
152
Chapter5DirectPhysicalDamageGeneralBuildingStock
used to estimate damage due to ground shaking. Building damage functions for ground failure
are described in Section 5.5.

8. Lifelines-
Utility
Systems
4. Ground Motion 4. Ground Failure
Direct Physical
Damage
6. Essential and
High Potential
Loss Facilities
12. Debris 10. Fire 15. Economic 14. Shelter 9. Inundation 11. HazMat
16. Indirect
Economic
Losses
Potential Earth Science Hazards
Direct Economic/
Social Losses
Induced Physical
Damage
7. Lifelines-
Transportation
Systems
5. General
Building
Stock
13. Casualities

Flowchart5.1BuildingDamageRelationshiptoOtherComponentsoftheMethodology
153
HazusMHTechnicalManual
Section 5.6 describes implementation of ground shaking damage functions (including
development of damping-modified demand spectra) and the calculation of the probability of
combined ground shaking and ground failure damage.

The methods described in this chapter may also be used by seismic/structural engineering experts
to modify default damage functions (based on improved knowledge of building types, their
structural properties and design vintage). Guidance for expert users is provided in Section 5.7

5.1.2 Input Requirements and Output Information



Input required to estimate building damage using fragility and capacity curves includes the
following two items:

- model building type (including height) and seismic design level that represents the building
(or group of buildings) of interest, and

- response spectrum (or PGA, for lifeline buildings, and PGD for ground failure evaluation) at
the buildings site or at the centroid of the census tract area where the building (or group of
buildings) is located.

Typically, the model building type is not known for each building and must be determined from
the inventory of facilities using the relationship of building type and occupancy, described in
Chapter 3. The response spectrum, PGA and PGD at the building site (or census tract centroid)
are PESH outputs, described in Chapter 4.

The output of fragility curves is an estimate of the cumulative probability of being in, or
exceeding, each damage state for the given level of ground shaking (or ground failure). Discrete
damage state probabilities are created using cumulative damage probabilities, as described in
Section 5.6. Discrete damage state probabilities for model building types and occupancy classes
are the outputs of the building damage module. These outputs are used directly as inputs to
induced physical damage and direct economic and social loss modules, as shown in Flowchart
5.1. While the fragility and capacity curves are applicable, in theory, to a single building as well
as to all buildings of given type, they are more reliable as predictors of damage for large, rather
than small, population groups. They should not be considered reliable for prediction of damage
to a specific facility without confirmation by a seismic/structural engineering expert.
154
Chapter5DirectPhysicalDamageGeneralBuildingStock

5.1.3 Form of Damage Functions

Building damage functions are in the form of lognormal fragility curves that relate the probability
of being in, or exceeding, a building damage state to for a given PESH demand parameter (e.g.,
response spectrum displacement). Figure 5.1 provides an example of fragility curves for the four
damage states used in this methodology.
Each fragility curve is defined by a median value of the PESH demand parameter (i.e., either
spectral displacement, spectral acceleration, PGA or PGD) that corresponds to the threshold of
the damage state and by the variability associated with that damage state. For example, the
spectral displacement, S
d
, that defines the threshold of a particular damage state (ds) is assumed
to be distributed by:

ds ds d d
S S c - =
,
(51)
where: S
d,ds
is the median value of spectral displacement of damage state, ds,
and
c
ds
is a lognormal random variable with unit median value and
logarithmic standard deviation, |
ds
.

155
HazusMHTechnicalManual
Figure5.1ExampleFragilityCurvesforSlight,Moderate,ExtensiveandCompleteDamage.
In a more general formulation of fragility curves, the lognormal standard deviation, |, has been
expressed in terms of the randomness and uncertainty components of variability, |
R
and |
U
,
[Kennedy, et. al., 1980]. Since it is not considered practical to separate uncertainty from
randomness, the combined random variable term, |, is used to develop a composite best-
estimate fragility curve. This approach is similar to that used to develop fragility curves for the
FEMA-sponsored study of consequences of large earthquakes on six cities of the Mississippi
Valley region [Allen & Hoshall, et al., 1985].

The conditional probability of being in, or exceeding, a particular damage state, ds, given the
spectral displacement, S
d
, (or other PESH parameter) is defined by the function:


| |
P S
1
ln
S
S
d
d
,
ds
ds
d ds
=
|
\

|
.
|

(
u
|
(5-2)

where: ds d S , isthemedianvalueofspectraldisplacementatwhichthebuilding
reachesthethresholdofdamagestate,ds,
|ds isthestandard deviation of the natural logarithm of spectral
displacementfordamagestate,ds,and
u isthestandardnormalcumulativedistributionfunction.
Median spectral displacement (or acceleration) values and the total variability are developed for
each of the model building types and damage states of interest by the combination of performance
data (from tests of building elements), earthquake experience data, expert opinion and judgment.

In general, the total variability of each damage state, |
ds
, is modeled by the combination of
following three contributors to damage variability:

- uncertaintyinthedamagestatethreshold,
- variability in the capacity (response) properties of the model building type
ofinterest,and
156
Chapter5DirectPhysicalDamageGeneralBuildingStock
- uncertainty in response due to the spatial variability of ground motion
demand.
Each of these three contributors to damage state variability is assumed to be lognormally
distributed random variables.

The fragility curves are driven by a PESH parameter. For ground failure, the PESH parameter
used to drive fragility curves is permanent ground displacement (PGD). For ground shaking, the
PESH parameter used to drive building fragility curves is peak spectral response (either
displacement or acceleration). Peak ground acceleration (PGA), rather than peak spectral
displacement, is used to evaluate ground shaking-induced structural damage to buildings that are
components of lifelines (see Section 5.4.4). Peak spectral response varies significantly for
buildings that have different response properties (e.g., tall, flexible buildings will displace more
than short, stiff buildings). Therefore, determination of peak spectral displacement requires
knowledge of the buildings response properties.

Building response is characterized by building capacity curves. These curves describe the push-
over displacement of each building type and seismic design level as a function of laterally-applied
earthquake load. The Methodology uses a technique, similar to the capacity spectrum method
[Mahaney, et. al., 1993], to estimate peak building response as the intersection of the building
capacity curve and the response spectrum of PESH shaking demand at the buildings location
(demand spectrum). The capacity spectrum method is one of the two nonlinear static analysis
methods described in theNEHRP Guidelines for the Seismic Rehabilitation of Buildings [FEMA,
1996a] and developed more extensively in Seismic Evaluation and Retrofit of Concrete Buildings
[SSC, 1996].

The demand spectrum is the 5%-damped PESH input spectrum reduced for higher levels of
effective damping (e.g., effective damping includes both elastic damping and hysteretic damping
associated with post-yield cyclic response of the building). Figure 5.2 illustrates the intersection
of a typical building capacity curve and a typical demand spectrum (reduced for effective
damping greater than 5% of critical). Design-, yield- and ultimate-capacity points define the
shape of building capacity curves. Peak building response (either spectral displacement or
spectral acceleration) at the point of intersection of the capacity curve and demand spectrum is
the parameter used with fragility curves to estimate damage state probabilities (see also Section
5.6.2.2).

157
HazusMHTechnicalManual

Figure5.2ExampleBuildingCapacityCurveandDemandSpectrum.

5.2 DescriptionofModelBuildingTypes

Table 5.1 lists the 36 model building types that are used by the Methodology. These model
building types are based on the classification system of FEMA 178, NEHRP Handbook for the
Seismic Evaluation of Existing Buildings [FEMA, 1992]. In addition, the methodology breaks
downFEMA178classesintoheightranges,andalsoincludesmobilehomes.

Table5.1ModelBuildingTypes
Height
No. Label Description Range Typical
Name Stories Stories Feet
1
2
W1
W2
Wood,LightFrame(s5,000sq.ft.)
Wood,CommercialandIndustrial(>5,000
sq.ft.)
12
All
1
2
14
24
3
4
S1L
S1M
SteelMomentFrame LowRise
MidRise
13
47
2
5
24
60
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 1 2 3 4 5 6 7 8 9 10
Spectral Displacement (inches)
S
p
e
c
t
r
a
l

A
c
c
e
l
e
r
a
t
i
o
n

(
g
'
s
)
PESH Input Spectrum
(5% Damping)
Demand Spectrum
(Damping > 5%)
Capacity
Curve
Ultimate Capacity
Design Capacity
Yield Capacity
S
a
S
d
158
Chapter5DirectPhysicalDamageGeneralBuildingStock
5 S1H HighRise 8+ 13 156
6
7
8
S2L
S2M
S2H
SteelBracedFrame LowRise
MidRise
HighRise
13
47
8+
2
5
13
24
60
156
9 S3 SteelLightFrame All 1 15
10
11
12
S4L
S4M
S4H
SteelFramewithCastinPlaceConcrete
ShearWalls
LowRise
MidRise
HighRise
13
47
8+
2
5
13
24
60
156
13
14
15
S5L
S5M
S5H
SteelFramewithUnreinforcedMasonry
InfillWalls
LowRise
MidRise
HighRise
13
47
8+
2
5
13
24
60
156
16
17
18
C1L
C1M
C1H
ConcreteMomentFrame LowRise
MidRise
HighRise
13
47
8+
2
5
12
20
50
120
19
20
21
C2L
C2M
C2H
ConcreteShearWalls LowRise
MidRise
HighRise
13
47
8+
2
5
12
20
50
120
22
23
24
C3L
C3M
C3H
ConcreteFramewithUnreinforced
MasonryInfillWalls
LowRise
MidRise
HighRise
13
47
8+
2
5
12
20
50
120
25 PC1 PrecastConcreteTiltUpWalls All 1 15
26
27
28
PC2L
PC2M
PC2H
PrecastConcreteFrameswithConcrete
ShearWalls
LowRise
MidRise
HighRise
13
47
8+
2
5
12
20
50
120
29
30
RM1L
RM1M
ReinforcedMasonryBearingWallswith
WoodorMetalDeckDiaphragms
LowRise
MidRise
13
4+
2
5
20
50
159
HazusMHTechnicalManual
31
32
33
RM2L
RM2M
RM2H
ReinforcedMasonryBearingWallswith
PrecastConcreteDiaphragms
LowRise
MidRise
HighRise
13
47
8+
2
5
12
20
50
120
34
35
URML
URMM
UnreinforcedMasonryBearingWalls LowRise
MidRise
12
3+
1
3
15
35
36 MH MobileHomes All 1 10
5.2.1 StructuralSystems

A general description of each of the 16 structural systems of model building types is given in the
following sections.

Wood, Light Frame (W1):
These are typically single-family or small, multiple-family dwellings of not more than 5,000
square feet of floor area. The essential structural feature of these buildings is repetitive framing
by wood rafters or joists on wood stud walls. Loads are light and spans are small. These
buildings may have relatively heavy masonry chimneys and may be partially or fully covered
with masonry veneer. Most of these buildings, especially the single-family residences, are not
engineered but constructed in accordance with conventional construction provisions of building
codes. Hence, they usually have the components of a lateral-force-resisting system even though it
may be incomplete. Lateral loads are transferred by diaphragms to shear walls. The diaphragms
are roof panels and floors that may be sheathed with sawn lumber, plywood or fiberboard
sheathing. Shear walls are sheathed with boards, stucco, plaster, plywood, gypsum board, particle
board, or fiberboard, or interior partition walls sheathed with plaster or gypsum board.

Wood, Greater than 5,000 Sq. Ft. (W2):
These buildings are typically commercial or industrial buildings, or multi-family residential
buildings with a floor area greater than 5,000 square feet. These buildings include structural
systems framed by beams or major horizontally spanning members over columns. These
horizontal members may be glue-laminated (glu-lam) wood, solid-sawn wood beams, or wood
trusses, or steel beams or trusses. Lateral loads usually are resisted by wood diaphragms and
exterior walls sheathed with plywood, stucco, plaster, or other paneling. The walls may have
diagonal rod bracing. Large openings for stores and garages often require post-and-beam
framing. Lateral load resistance on those lines may be achieved with steel rigid frames (moment
frames) or diagonal bracing.
1510
Chapter5DirectPhysicalDamageGeneralBuildingStock

Steel Moment Frame (S1):
These buildings have a frame of steel columns and beams. In some cases, the beam-column
connections have very small moment resisting capacity but, in other cases, some of the beams and
columns are fully developed as moment frames to resist lateral forces. Usually the structure is
concealed on the outside by exterior nonstructural walls, which can be of almost any material
(curtain walls, brick masonry, or precast concrete panels), and on the inside by ceilings and
column furring. Diaphragms transfer lateral loads to moment-resisting frames. The diaphragms
can be almost any material. The frames develop their stiffness by full or partial moment
connections. The frames can be located almost anywhere in the building. Usually the columns
have their strong directions oriented so that some columns act primarily in one direction while the
others act in the other direction. Steel moment frame buildings are typically more flexible than
shear wall buildings. This low stiffness can result in large interstory drifts that may lead to
relatively greater nonstructural damage.

Steel Braced Frame (S2):
These buildings are similar to steel moment frame buildings except that the vertical components
of the lateral-force-resisting system are braced frames rather than moment frames.

Steel Light Frame (S3):
These buildings are pre-engineered and prefabricated with transverse rigid frames. The roof and
walls consist of lightweight panels, usually corrugated metal. The frames are designed for
maximum efficiency, often with tapered beam and column sections built up of light steel plates.
The frames are built in segments and assembled in the field with bolted joints. Lateral loads in
the transverse direction are resisted by the rigid frames with loads distributed to them by
diaphragm elements, typically rod-braced steel roof framing bays. Tension rod bracing typically
resists loads in the longitudinal direction.

Steel Frame with Cast-In-Place Concrete Shear Walls (S4):
The shear walls in these buildings are cast-in-place concrete and may be bearing walls. The steel
frame is designed for vertical loads only. Diaphragms of almost any material transfer lateral
loads to the shear walls. The steel frame may provide a secondary lateral-force-resisting system
depending on the stiffness of the frame and the moment capacity of the beam-column
connections. In modern dual systems, the steel moment frames are designed to work together
with the concrete shear walls.

1511
HazusMHTechnicalManual
Steel Frame with Unreinforced Masonry Infill Walls (S5):
This is one of the older types of buildings. The infill walls usually are offset from the exterior
frame members, wrap around them, and present a smooth masonry exterior with no indication of
the frame. Solidly infilled masonry panels, when they fully engage the surrounding frame
members (i.e. lie in the same plane), may provide stiffness and lateral load resistance to the
structure.

Reinforced Concrete Moment Resisting Frames (C1):
These buildings are similar to steel moment frame buildings except that the frames are reinforced
concrete. There are a large variety of frame systems. Some older concrete frames may be
proportioned and detailed such that brittle failure of the frame members can occur in earthquakes
leading to partial or full collapse of the buildings. Modern frames in zones of high seismicity are
proportioned and detailed for ductile behavior and are likely to undergo large deformations
during an earthquake without brittle failure of frame members and collapse.

Concrete Shear Walls (C2):
The vertical components of the lateral-force-resisting system in these buildings are concrete shear
walls that are usually bearing walls. In older buildings, the walls often are quite extensive and the
wall stresses are low but reinforcing is light. In newer buildings, the shear walls often are limited
in extent, generating concerns about boundary members and overturning forces.

Concrete Frame Buildings with Unreinforced Masonry Infill Walls (C3):
These buildings are similar to steel frame buildings with unreinforced masonry infill walls except
that the frame is of reinforced concrete. In these buildings, the shear strength of the columns,
after cracking of the infill, may limit the semi-ductile behavior of the system.

Precast Concrete Tilt-Up Walls (PC1):
These buildings have a wood or metal deck roof diaphragm, which often is very large, that
distributes lateral forces to precast concrete shear walls. The walls are thin but relatively heavy
while the roofs are relatively light. Older or non-seismic-code buildings often have inadequate
connections for anchorage of the walls to the roof for out-of-plane forces, and the panel
connections often are brittle. Tilt-up buildings usually are one or two stories in height. Walls can
have numerous openings for doors and windows of such size that the wall looks more like a
frame than a shear wall.

1512
Chapter5DirectPhysicalDamageGeneralBuildingStock
Precast Concrete Frames with Concrete Shear Walls (PC2):
These buildings contain floor and roof diaphragms typically composed of precast concrete
elements with or without cast-in-place concrete topping slabs. Precast concrete girders and
columns support the diaphragms. The girders often bear on column corbels. Closure strips
between precast floor elements and beam-column joints usually are cast-in-place concrete.
Welded steel inserts often are used to interconnect precast elements. Precast or cast-in-place
concrete shear walls resist lateral loads. For buildings with precast frames and concrete shear
walls to perform well, the details used to connect the structural elements must have sufficient
strength and displacement capacity; however, in some cases, the connection details between the
precast elements have negligible ductility.

Reinforced Masonry Bearing Walls with Wood or Metal Deck Diaphragms (RM1):
These buildings have perimeter bearing walls of reinforced brick or concrete-block masonry.
These walls are the vertical elements in the lateral-force-resisting system. The floors and roofs
are framed with wood joists and beams either with plywood or braced sheathing, the latter either
straight or diagonally sheathed, or with steel beams with metal deck with or without concrete fill.
Interior wood posts or steel columns support wood floor framing; steel columns support steel
beams.

Reinforced Masonry Bearing Walls with Precast Concrete Diaphragms (RM2):
These buildings have bearing walls similar to those of reinforced masonry bearing wall structures
with wood or metal deck diaphragms, but the roof and floors are composed of precast concrete
elements such as planks or tee-beams and the precast roof and floor elements are supported on
interior beams and columns of steel or concrete (cast-in-place or precast). The precast horizontal
elements often have a cast-in-place topping.

Unreinforced Masonry Bearing Walls (URM):
These buildings include structural elements that vary depending on the buildings age and, to a
lesser extent, its geographic location. In buildings built before 1900, the majority of floor and
roof construction consists of wood sheathing supported by wood framing. In large multistory
buildings, the floors are cast-in-place concrete supported by the unreinforced masonry walls
and/or steel or concrete interior framing. In unreinforced masonry constructed after 1950 (outside
California) wood floors usually have plywood rather than board sheathing. In regions of lower
seismicity, buildings of this type constructed more recently can include floor and roof framing
that consists of metal deck and concrete fill supported by steel framing elements. The perimeter
walls, and possibly some interior walls, are unreinforced masonry. The walls may or may not be
anchored to the diaphragms. Ties between the walls and diaphragms are more common for the
1513
HazusMHTechnicalManual
bearing walls than for walls that are parallel to the floor framing. Roof ties usually are less
common and more erratically spaced than those at the floor levels. Interior partitions that
interconnect the floors and roof can reduce diaphragm displacements.

Mobile Homes (MH):
These are prefabricated housing units that are trucked to the site and then placed on isolated piers,
jack stands, or masonry block foundations (usually without any positive anchorage). Floors and
roofs of mobile homes usually are constructed with plywood and outside surfaces are covered
with sheet metal.

5.2.2 NonstructuralComponents
Nonstructural components include a large variety of different architectural, mechanical and
electrical components (e.g., components listed in the NEHRP seismic design provisions for new
buildings [FEMA, 1997a]). Contents of the buildings are treated as a separate category.
Nonstructural components are grouped as either "drift-sensitive" or "acceleration-sensitive"
components, in order to assess their damage due to an earthquake. Damage to drift-sensitive
nonstructural components is primarily a function of interstory drift; damage to acceleration-
sensitive nonstructural components and building contents is primarily a function of floor
acceleration. Table 5.2 lists typical nonstructural components and building contents, and
identifies each item as drift-sensitive or acceleration sensitive.

Anchorage/bracing of nonstructural components improves earthquake performance of most
components although routine or typical anchorage/bracing provides only limited damage
protection. It is assumed that typical nonstructural components and building contents have
limited anchorage/bracing. Exceptions, such as special anchorage/bracing requirements for
nonstructural components and contents of hospitals are addressed in Chapter 6. Nonstructural
damage evaluation is dependent upon the response and performance of structural components, as
well as being influenced by characteristics of nonstructural components themselves.
Nonstructural damage simplifying assumptions are outlined in the following sections.
1514
Chapter5DirectPhysicalDamageGeneralBuildingStock
Table 5.2 List of Typical Nonstructural Components and Contents of Buildings

Type Item
Drift
Sensitive*
Acceleration
Sensitive*
Architectural NonbearingWalls/Partitions -
CantileverElementsandParapets -
ExteriorWallPanels -
VeneerandFinishes -
Penthouses -
RacksandCabinets -
AccessFloors -
AppendagesandOrnaments -
Mechanical GeneralMechanical(boilers,etc.) -
and ManufacturingandProcessMachinery -
Electrical PipingSystems -
StorageTanksandSpheres -
HVACSystems(chillers,ductwork,etc.) -
Elevators -
TrussedTowers -
GeneralElectrical(switchgear,ducts,etc.) -
LightingFixtures -
Contents FileCabinets,Bookcases,etc. -
OfficeEquipmentandFurnishings -
Computer/CommunicationEquipment -
NonpermanentManufacturingEquipment -
Manufacturing/StorageInventory -
ArtandotherValuableObjects -
* Soliddotsindicateprimarycauseofdamage,opendotsindicatesecondarycauseofdamage

1515
HazusMHTechnicalManual
5.3 DescriptionofBuildingDamageStates

The results of damage estimation methods described in this chapter (i.e., damage predictions for
model building types for a given level of ground shaking) are used in other modules of the
methodology to estimate: (1) casualties due to structural damage, including fatalities, (2)
monetary losses due to building damage (i.e. cost of repairing or replacing damaged buildings and
their contents); (3) monetary losses resulting from building damage and closure (e.g., losses due
to business interruption); (4) social impacts (e.g., loss of shelter); and, (5) other economic and
social impacts.

The building damage predictions may also be used to study expected damage patterns in a given
region for different scenario earthquakes (e.g., to identify the most vulnerable building types, or
the areas expected to have the most damaged buildings).

In order to meet the needs of such broad purposes, damage predictions must allow the user to
glean the nature and extent of the physical damage to a building type from the damage prediction
output so that life-safety, societal functional and monetary losses which result from the damage
can be estimated. Building damage can best be described in terms of its components (beams,
columns, walls, ceilings, piping, HVAC equipment, etc.). For example, such component damage
descriptions as shear walls are cracked, ceiling tiles fell, diagonal bracing buckled, wall
panels fell out, etc. used together with such terms as some and most would be sufficient to
describe the nature and extent of overall building damage.

Damage to nonstructural components of buildings (i.e., architectural components, such as
partition walls and ceilings, and building mechanical/electrical systems) primarily affects
monetary and societal functional losses and generates numerous casualties of mostly light-to-
moderate severity. Damage to structural components (i.e., the gravity and lateral-load-resisting
systems) of buildings, Hazard mitigation measures are different for these two categories of
building components as well. Hence, it is desirable to separately estimate structural and
nonstructural damage.

Building damage varies from none to complete as a continuous function of building
deformations (building response). Wall cracks may vary from invisible or hairline cracks to
cracks of several inches wide. Generalized ranges of damage are used by the Methodology to
describe structural and nonstructural damage, since it is not practical to describe building damage
as a continuous function.

1516
Chapter5DirectPhysicalDamageGeneralBuildingStock
The Methodology predicts a structural and nonstructural damage state in terms of one of four
ranges of damage or damage states: Slight, Moderate, Extensive, and Complete. For example,
the Slight damage state extends from the threshold of Slight damage up to the threshold of
Moderate damage. General descriptions of these damage states are provided for all model
building types with reference to observable damage incurred by structural (Section 5.3.1) and
nonstructural building components (Section 5.3.2). Damage predictions resulting from this
physical damage estimation method are then expressed in terms of the probability of a building
being in any of these four damage states.

5.3.1 Structural Damage



Descriptions for Slight, Moderate, Extensive, and Complete structural damage states for the 16
basic model building types are provided below. For estimating casualties, the descriptions of
Complete damage include the fraction of the total floor area of each model building type that is
likely to collapse. Collapse fractions are based on judgment and limited earthquake data
considering the material and construction of different model building types.

It is noted that in some cases the structural damage is not directly observable because the
structural elements are inaccessible or not visible due to architectural finishes or fireproofing.
Hence, these structural damage states are described, when necessary, with reference to certain
effects on nonstructural elements that may be indicative of the structural damage state of concern.
Small cracks are assumed, throughout this section, to be visible cracks with a maximum width of
less than 1/8. Cracks wider than 1/8 are referred to as large cracks.

Wood,LightFrame(W1):
Slight Structural Damage: Small plaster or gypsum-board cracks at corners of door and window
openings and wall-ceiling intersections; small cracks in masonry chimneys and masonry veneer.
Moderate Structural Damage: Large plaster or gypsum-board cracks at corners of door and
window openings; small diagonal cracks across shear wall panels exhibited by small cracks in
stucco and gypsum wall panels; large cracks in brick chimneys; toppling of tall masonry
chimneys.
Extensive Structural Damage: Large diagonal cracks across shear wall panels or large cracks at
plywood joints; permanent lateral movement of floors and roof; toppling of most brick chimneys;
cracks in foundations; splitting of wood sill plates and/or slippage of structure over foundations;
partial collapse of room-over-garage or other soft-story configurations; small foundations
cracks.
1517
HazusMHTechnicalManual
Complete Structural Damage: Structure may have large permanent lateral displacement, may
collapse, or be in imminent danger of collapse due to cripple wall failure or the failure of the
lateral load resisting system; some structures may slip and fall off the foundations; large
foundation cracks. Approximately 3% of the total area of W1 buildings with Complete damage is
expected to be collapsed.

Wood,CommercialandIndustrial(W2):
Slight Structural Damage: Small cracks at corners of door and window openings and wall-
ceiling intersections; small cracks on stucco and plaster walls. Some slippage may be observed at
bolted connections.
Moderate Structural Damage: Larger cracks at corners of door and window openings; small
diagonal cracks across shear wall panels exhibited by cracks in stucco and gypsum wall panels;
minor slack (less than 1/8 extension) in diagonal rod bracing requiring re-tightening; minor
lateral set at store fronts and other large openings; small cracks or wood splitting may be
observed at bolted connections.
Extensive Structural Damage: Large diagonal cracks across shear wall panels; large slack in
diagonal rod braces and/or broken braces; permanent lateral movement of floors and roof; cracks
in foundations; splitting of wood sill plates and/or slippage of structure over foundations; partial
collapse of soft-story configurations; bolt slippage and wood splitting at bolted connections.
Complete Structural Damage: Structure may have large permanent lateral displacement, may
collapse or be in imminent danger of collapse due to failed shear walls, broken brace rods or
failed framing connections; it may fall its foundations; large cracks in the foundations.
Approximately 3% of the total area of W2 buildings with Complete damage is expected to be
collapsed.

SteelMomentFrame(S1):
Slight Structural Damage: Minor deformations in connections or hairline cracks in few welds.
Moderate Structural Damage: Some steel members have yielded exhibiting observable
permanent rotations at connections; few welded connections may exhibit major cracks through
welds or few bolted connections may exhibit broken bolts or enlarged bolt holes.
Extensive Structural Damage: Most steel members have exceeded their yield capacity, resulting
in significant permanent lateral deformation of the structure. Some of the structural members or
connections may have exceeded their ultimate capacity exhibited by major permanent member
rotations at connections, buckled flanges and failed connections. Partial collapse of portions of
structure is possible due to failed critical elements and/or connections.
1518
Chapter5DirectPhysicalDamageGeneralBuildingStock
Complete Structural Damage: Significant portion of the structural elements have exceeded their
ultimate capacities or some critical structural elements or connections have failed resulting in
dangerous permanent lateral displacement, partial collapse or collapse of the building.
Approximately 8%(low-rise), 5%(mid-rise) or 3%(high-rise) of the total area of S1 buildings with
Complete damage is expected to be collapsed.

SteelBracedFrame(S2):
Slight Structural Damage: Few steel braces have yielded which may be indicated by minor
stretching and/or buckling of slender brace members; minor cracks in welded connections; minor
deformations in bolted brace connections.
Moderate Structural Damage: Some steel braces have yielded exhibiting observable stretching
and/or buckling of braces; few braces, other members or connections have indications of reaching
their ultimate capacity exhibited by buckled braces, cracked welds, or failed bolted connections.
Extensive Structural Damage: Most steel brace and other members have exceeded their yield
capacity, resulting in significant permanent lateral deformation of the structure. Some structural
members or connections have exceeded their ultimate capacity exhibited by buckled or broken
braces, flange buckling, broken welds, or failed bolted connections. Anchor bolts at columns
may be stretched. Partial collapse of portions of structure is possible due to failure of critical
elements or connections.
Complete Structural Damage: Most the structural elements have reached their ultimate
capacities or some critical members or connections have failed resulting in dangerous permanent
lateral deflection, partial collapse or collapse of the building. Approximately 8%(low-rise),
5%(mid-rise) or 3%(high-rise) of the total area of S2 buildings with Complete damage is
expected to be collapsed.

SteelLightFrame(S3):
These structures are mostly single story structures combining rod-braced frames in one direction
and moment frames in the other. Due to repetitive nature of the structural systems, the type of
damage to structural members is expected to be rather uniform throughout the structure.
Slight Structural Damage: Few steel rod braces have yielded which may be indicated by minor
sagging of rod braces. Minor cracking at welded connections or minor deformations at bolted
connections of moment frames may be observed.
Moderate Structural Damage: Most steel braces have yielded exhibiting observable
significantly sagging rod braces; few brace connections may be broken. Some weld cracking may
be observed in the moment frame connections.
1519
HazusMHTechnicalManual
Extensive Structural Damage: Significant permanent lateral deformation of the structure due to
broken brace rods, stretched anchor bolts and permanent deformations at moment frame
members. Some screw or welded attachments of roof and wall siding to steel framing may be
broken. Some purlin and girt connections may be broken.
Complete Structural Damage: Structure is collapsed or in imminent danger of collapse due to
broken rod bracing, failed anchor bolts or failed structural members or connections.
Approximately 3% of the total area of S3 buildings with Complete damage is expected to be
collapsed.

SteelFramewithCastInPlaceConcreteShearWalls(S4):
This is a composite structural system where primary lateral-force-resisting system is the
concrete shear walls. Hence, slight, Moderate and Extensive damage states are likely to be
determined by the shear walls while the collapse damage state would be determined by the failure
of the structural frame.
Slight Structural Damage: Diagonal hairline cracks on most concrete shear wall surfaces; minor
concrete spalling at few locations.
Moderate Structural Damage: Most shear wall surfaces exhibit diagonal cracks; some of the
shear walls have exceeded their yield capacities exhibited by larger diagonal cracks and concrete
spalling at wall ends.
Extensive Structural Damage: Most concrete shear walls have exceeded their yield capacities;
few walls have reached or exceeded their ultimate capacity exhibited by large through-the wall
diagonal cracks, extensive spalling around the cracks and visibly buckled wall reinforcement.
Partial collapse may occur due to failed connections of steel framing to concrete walls. Some
damage may be observed in steel frame connections.
Complete Structural Damage: Structure may be in danger of collapse or collapse due to total
failure of shear walls and loss of stability of the steel frames. Approximately 8%(low-rise),
5%(mid-rise) or 3%(high-rise) of the total area of S4 buildings with Complete damage is
expected to be collapsed.

SteelFramewithUnreinforcedMasonryInfillWalls(S5):
This is a composite structural system where the initial lateral resistance is provided by the infill
walls. Upon cracking of the infills, further lateral resistance is provided by the steel frames
braced by the infill walls acting as diagonal compression struts. Collapse of the structure
results when the infill walls disintegrate (due to compression failure of the masonry struts) and
the steel frame loses its stability.

1520
Chapter5DirectPhysicalDamageGeneralBuildingStock
Slight Structural Damage: Diagonal (sometimes horizontal) hairline cracks on most infill walls;
cracks at frame-infill interfaces.
Moderate Structural Damage: Most infill wall surfaces exhibit larger diagonal or horizontal
cracks; some walls exhibit crushing of brick around beam-column connections.
Extensive Structural Damage: Most infill walls exhibit large cracks; some bricks may be
dislodged and fall; some infill walls may bulge out-of-plane; few walls may fall off partially or
fully; some steel frame connections may have failed. Structure may exhibit permanent lateral
deformation or partial collapse due to failure of some critical members.
Complete Structural Damage: Structure is collapsed or in danger of imminent collapse due to
total failure of many infill walls and loss of stability of the steel frames. . Approximately
8%(low-rise), 5%(mid-rise) or 3%(high-rise) of the total area of S5 buildings with Complete
damage is expected to be collapsed.

ReinforcedConcreteMomentResistingFrames(C1):
Slight Structural Damage: Flexural or shear type hairline cracks in some beams and columns
near joints or within joints.
Moderate Structural Damage: Most beams and columns exhibit hairline cracks. In ductile
frames some of the frame elements have reached yield capacity indicated by larger flexural cracks
and some concrete spalling. Nonductile frames may exhibit larger shear cracks and spalling.
Extensive Structural Damage: Some of the frame elements have reached their ultimate capacity
indicated in ductile frames by large flexural cracks, spalled concrete and buckled main
reinforcement; nonductile frame elements may have suffered shear failures or bond failures at
reinforcement splices, or broken ties or buckled main reinforcement in columns which may result
in partial collapse.
Complete Structural Damage: Structure is collapsed or in imminent danger of collapse due to
brittle failure of nonductile frame elements or loss of frame stability. Approximately 13%(low-
rise), 10%(mid-rise) or 5%(high-rise) of the total area of C1 buildings with Complete damage is
expected to be collapsed.

ConcreteShearWalls(C2):
Slight Structural Damage: Diagonal hairline cracks on most concrete shear wall surfaces; minor
concrete spalling at few locations.
Moderate Structural Damage: Most shear wall surfaces exhibit diagonal cracks; some shear
walls have exceeded yield capacity indicated by larger diagonal cracks and concrete spalling at
wall ends.
1521
HazusMHTechnicalManual
Extensive Structural Damage: Most concrete shear walls have exceeded their yield capacities;
some walls have exceeded their ultimate capacities indicated by large, through-the-wall diagonal
cracks, extensive spalling around the cracks and visibly buckled wall reinforcement or rotation of
narrow walls with inadequate foundations. Partial collapse may occur due to failure of nonductile
columns not designed to resist lateral loads.
Complete Structural Damage: Structure has collapsed or is in imminent danger of collapse due
to failure of most of the shear walls and failure of some critical beams or columns.
Approximately 13%(low-rise), 10%(mid-rise) or 5%(high-rise) of the total area of C2 buildings
with Complete damage is expected to be collapsed.

ConcreteFrameBuildingswithUnreinforcedMasonryInfillWalls(C3):
This is a composite structural system where the initial lateral resistance is provided by the infill
walls. Upon cracking of the infills, further lateral resistance is provided by the concrete frame
braced by the infill acting as diagonal compression struts. Collapse of the structure results
when the infill walls disintegrate (due to compression failure of the masonry struts) and the
frame loses stability, or when the concrete columns suffer shear failures due to reduced effective
height and the high shear forces imposed on them by the masonry compression struts.

Slight Structural Damage: Diagonal (sometimes horizontal) hairline cracks on most infill walls;
cracks at frame-infill interfaces.
Moderate Structural Damage: Most infill wall surfaces exhibit larger diagonal or horizontal
cracks; some walls exhibit crushing of brick around beam-column connections. Diagonal shear
cracks may be observed in concrete beams or columns.
Extensive Structural Damage: Most infill walls exhibit large cracks; some bricks may dislodge
and fall; some infill walls may bulge out-of-plane; few walls may fall partially or fully; few
concrete columns or beams may fail in shear resulting in partial collapse. Structure may exhibit
permanent lateral deformation.
Complete Structural Damage: Structure has collapsed or is in imminent danger of collapse due
to a combination of total failure of the infill walls and nonductile failure of the concrete beams
and columns. Approximately 15%(low-rise), 13%(mid-rise) or 5%(high-rise) of the total area of
C3 buildings with Complete damage is expected to be collapsed.

PrecastConcreteTiltUpWalls(PC1):
Slight Structural Damage: Diagonal hairline cracks on concrete shear wall surfaces; larger
cracks around door and window openings in walls with large proportion of openings; minor
concrete spalling at few locations; minor separation of walls from the floor and roof diaphragms;
1522
Chapter5DirectPhysicalDamageGeneralBuildingStock
hairline cracks around metal connectors between wall panels and at connections of beams to
walls.
Moderate Structural Damage: Most wall surfaces exhibit diagonal cracks; larger cracks in
walls with door or window openings; few shear walls have exceeded their yield capacities
indicated by larger diagonal cracks and concrete spalling. Cracks may appear at top of walls near
panel intersections indicating chord yielding. Some walls may have visibly pulled away from
the roof. Some welded panel connections may have been broken, indicated by spalled concrete
around connections. Some spalling may be observed at the connections of beams to walls.
Extensive Structural Damage: In buildings with relatively large area of wall openings most
concrete shear walls have exceeded their yield capacities and some have exceeded their ultimate
capacities indicated by large, through-the-wall diagonal cracks, extensive spalling around the
cracks and visibly buckled wall reinforcement. The plywood diaphragms may exhibit cracking
and separation along plywood joints. Partial collapse of the roof may result from the failure of
the wall-to-diaphragm anchorages sometimes with falling of wall panels.
Complete Structural Damage: Structure is collapsed or is in imminent danger of collapse due to
failure of the wall-to-roof anchorages, splitting of ledgers, or failure of plywood-to-ledger nailing;
failure of beams connections at walls; failure of roof or floor diaphragms; or, failure of the wall
panels. Approximately 15% of the total area of PC1 buildings with Complete damage is expected
to be collapsed.

PrecastConcreteFrameswithConcreteShearWalls(PC2):
Slight Structural Damage: Diagonal hairline cracks on most shear wall surfaces; minor concrete
spalling at few connections of precast members.
Moderate Structural Damage: Most shear wall surfaces exhibit diagonal cracks; some shear
walls have exceeded their yield capacities indicated by larger cracks and concrete spalling at wall
ends; observable distress or movement at connections of precast frame connections, some failures
at metal inserts and welded connections.
Extensive Structural Damage: Most concrete shear walls have exceeded their yield capacities;
some walls may have reached their ultimate capacities indicated by large, through-the wall
diagonal cracks, extensive spalling around the cracks and visibly buckled wall reinforcement.
Some critical precast frame connections may have failed resulting partial collapse.
Complete Structural Damage: Structure has collapsed or is in imminent danger of collapse due
to failure of the shear walls and/or failures at precast frame connections. Approximately
15%(low-rise), 13%(mid-rise) or 10%(high-rise) of the total area of PC2 buildings with Complete
damage is expected to be collapsed.

1523
HazusMHTechnicalManual
ReinforcedMasonryBearingWallswithWoodorMetalDeckDiaphragms(RM1):
Slight Structural Damage: Diagonal hairline cracks on masonry wall surfaces; larger cracks
around door and window openings in walls with large proportion of openings; minor separation
of walls from the floor and roof diaphragms.
Moderate Structural Damage: Most wall surfaces exhibit diagonal cracks; some of the shear
walls have exceeded their yield capacities indicated by larger diagonal cracks. Some walls may
have visibly pulled away from the roof.
Extensive Structural Damage: In buildings with relatively large area of wall openings most
shear walls have exceeded their yield capacities and some of the walls have exceeded their
ultimate capacities indicated by large, through-the-wall diagonal cracks and visibly buckled wall
reinforcement. The plywood diaphragms may exhibit cracking and separation along plywood
joints. Partial collapse of the roof may result from failure of the wall-to-diaphragm anchorages or
the connections of beams to walls.
Complete Structural Damage: Structure has collapsed or is in imminent danger of collapse due
to failure of the wall anchorages or due to failure of the wall panels. Approximately 13%(low-
rise) or 10%(mid-rise) of the total area of RM1 buildings with Complete damage is expected to be
collapsed.

ReinforcedMasonryBearingWallswithPrecastConcreteDiaphragms(RM2):
Slight Structural Damage: Diagonal hairline cracks on masonry wall surfaces; larger cracks
around door and window openings in walls with large proportion of openings.
Moderate Structural Damage: Most wall surfaces exhibit diagonal cracks; some of the shear
walls have exceeded their yield capacities indicated by larger cracks.
Extensive Structural Damage: In buildings with relatively large area of wall openings most
shear walls have exceeded their yield capacities and some of the walls have exceeded their
ultimate capacities exhibited by large, through-the wall diagonal cracks and visibly buckled wall
reinforcement. The diaphragms may also exhibit cracking
Complete Structural Damage: Structure is collapsed or is in imminent danger of collapse due to
failure of the walls. Approximately 13%(low-rise), 10%(mid-rise) or 5%(high-rise) of the total
area of RM2 buildings with Complete damage is expected to be collapsed.

UnreinforcedMasonryBearingWalls(URM):
Slight Structural Damage: Diagonal, stair-step hairline cracks on masonry wall surfaces; larger
cracks around door and window openings in walls with large proportion of openings; movements
of lintels; cracks at the base of parapets.
1524
Chapter5DirectPhysicalDamageGeneralBuildingStock
Moderate Structural Damage: Most wall surfaces exhibit diagonal cracks; some of the walls
exhibit larger diagonal cracks; masonry walls may have visible separation from diaphragms;
significant cracking of parapets; some masonry may fall from walls or parapets.
Extensive Structural Damage: In buildings with relatively large area of wall openings most
walls have suffered extensive cracking. Some parapets and gable end walls have fallen. Beams
or trusses may have moved relative to their supports.
Complete Structural Damage: Structure has collapsed or is in imminent danger of collapse due
to in-plane or out-of-plane failure of the walls. Approximately 15% of the total area of URM
buildings with Complete damage is expected to be collapsed.

MobileHomes(MH):
Slight Structural Damage: Damage to some porches, stairs or other attached components.
Moderate Structural Damage: Major movement of the mobile home over its supports resulting
in some damage to metal siding and stairs and requiring resetting of the mobile home on its
supports.
Extensive Structural Damage: Mobile home has fallen partially off its supports, often severing
utility lines.
Complete Structural Damage: Mobile home has totally fallen off its supports; usually severing
utility lines, with steep jack stands penetrating through the floor. Approximately 3% of the total
area of MH buildings with Complete damage is expected to be collapsed.

5.3.2 NonstructuralDamage

Four damage states are used to describe nonstructural damage: Slight, Moderate, Extensive and
Complete nonstructural damage. Nonstructural damage is considered to be independent of the
structural model building type (i.e. partitions, ceilings, cladding, etc. are assumed to incur the
same damage when subjected to the same interstory drift or floor acceleration whether they are in
a steel frame building or in a concrete shear wall building), consequently, building-specific
damage state descriptions are not meaningful. Instead, general descriptions of nonstructural
damage states are provided for common nonstructural systems.

Damage to drift-sensitive nonstructural components is primarily a function of interstory drift (e.g.
full-height drywall partitions) while for acceleration-sensitive components (e.g. mechanical
equipment) damage is a function of the floor acceleration. Developing fragility curves for each
possible nonstructural component is not practicable for the purposes of regional loss estimation
1525
HazusMHTechnicalManual
and there is insufficient data to develop such fragility curves. Hence, in this methodology
nonstructural building components are grouped into drift-sensitive and acceleration-sensitive
component groups, and the damage functions estimated for each group are assumed to be
"typical" of it sub-components. Note, however, that damage depends on the anchorage/bracing
provided to the nonstructural components. Damageability characteristics of each group are
described by a set of fragility curves (see Subsection 5.4.3.3).

The type of nonstructural components in a given building is a function of the building occupancy-
use classification. For example, single-family residences would not have curtain wall panels,
suspended ceilings, elevators, etc. while these items would be found in an office building. Hence,
the relative values of nonstructural components in relation to the overall building replacement
value vary with type of occupancy. In Chapter 15, estimates of replacement cost breakdown
between structural building components for different occupancy/use related classifications are
provided; further breakdowns are provided by drift- and acceleration-sensitive nonstructural
components.

In the following, general descriptions of the four nonstructural damage states are described for
common nonstructural building components:

PartitionsWalls
Slight Nonstructural Damage: A few cracks are observed at intersections of walls and ceilings
and at corners of door openings.
Moderate Nonstructural Damage: Larger and more extensive cracks requiring repair and
repainting; some partitions may require replacement of gypsum board or other finishes.
Extensive Nonstructural Damage: Most of the partitions are cracked and a significant portion
may require replacement of finishes; some door frames in the partitions are also damaged and
require re-setting.
Complete Nonstructural Damage: Most partition finish materials and framing may have to be
removed and replaced; damaged studs repaired, and walls be refinished. Most door frames may
also have to be repaired and replaced.

SuspendedCeilings
Slight Nonstructural Damage: A few ceiling tiles have moved or fallen down.
1526
Chapter5DirectPhysicalDamageGeneralBuildingStock
Moderate Nonstructural Damage: Falling of tiles is more extensive; in addition the ceiling
support framing (T-bars) has disconnected and/or buckled at few locations; lenses have fallen off
of some light fixtures and a few fixtures have fallen; localized repairs are necessary.
Extensive Nonstructural Damage: The ceiling system exhibits extensive buckling, disconnected
t-bars and falling ceiling tiles; ceiling partially collapses at few locations and some light fixtures
fall; repair typically involves removal of most or all ceiling tiles.
Complete Nonstructural Damage: The ceiling system is buckled throughout and/or fallen and
requires complete replacement; many light fixtures fall.

ExteriorWallPanels
Slight Nonstructural Damage: Slight movement of the panels, requiring realignment.
Moderate Nonstructural Damage: The movements are more extensive; connections of panels to
structural frame are damaged requiring further inspection and repairs; some window frames may
need realignment
Extensive Nonstructural Damage: Most of the panels are cracked or otherwise damaged and
misaligned, and most panel connections to the structural frame are damaged requiring thorough
review and repairs; few panels fall or are in imminent danger of falling; some window panes are
broken and some pieces of glass have fallen.
Complete Nonstructural Damage: Most panels are severely damaged, most connections are
broken or severely damaged, some panels have fallen and most are in imminent danger of falling;
extensive glass breakage and falling.

Electrical-Mechanical Equipment, Piping, Ducts

Slight Nonstructural Damage: The most vulnerable equipment (e.g. unanchored or on spring
isolators) moves and damages attached piping or ducts.
Moderate Nonstructural Damage: Movements are larger and damage is more extensive; piping
leaks at few locations; elevator machinery and rails may require realignment
Extensive Nonstructural Damage: Equipment on spring isolators topples and falls; other
unanchored equipment slides or falls breaking connections to piping and ducts; leaks develop at
many locations; anchored equipment indicate stretched bolts or strain at anchorages.
Complete Nonstructural Damage: Equipment is damaged by sliding, overturning or failure of
their supports and is not operable; piping is leaking at many locations; some pipe and duct
1527
HazusMHTechnicalManual
supports have failed causing pipes and ducts to fall or hang down; elevator rails are buckled or
have broken supports and/or counterweights have derailed.

5.4 Building Damage Due to Ground Shaking

5.4.1 Overview

This section describes capacity and fragility curves used in the Methodology to estimate the
probability of Slight, Moderate, Extensive and Complete damage to general building stocks.
General building stock represents a population of a given model building type designed to either
High-Code, Moderate-Code, or Low-Code seismic standards, or not seismically designed,
referred to as to a Pre-Code buildings. Chapter 6 describes Special building damage functions for
estimating damage to hospitals and other essential facilities that are designed and constructed to
above average seismic standards.

Capacity curves and fragility curves for High-Code, Moderate-Code, Low-Code and Pre-Code
buildings are based on modern code (e.g., 1976 Uniform Building Code, 1985 NEHRP
Provisions, or later editions of these model codes). Design criteria for various seismic design
zones, as shown in Table 5.3. Additional description of seismic levels may be found in Section
5.7.

Table 5.3 Approximate Basis for Seismic Design Levels

SeismicDesignLevel SeismicZone
(UniformBuildingCode)
MapArea
(NEHRPProvisions)
HighCode
ModerateCode
LowCode
PreCode
4
2B
1
0
7
5
3
1

The capacity and fragility curves represent buildings designed and constructed to modern seismic
code provisions. Study areas (e.g., census tracts) of recent construction are appropriately
modeled using building damage functions with a seismic design level that corresponds to the
seismic zone or map area of the governing provisions. Older areas of construction, not
conforming to modern standards, should be modeled using a lower level of seismic design. For
1528
Chapter5DirectPhysicalDamageGeneralBuildingStock
example, in areas of high seismicity (e.g., coastal California), buildings of newer construction
(e.g., post-1973) are best represented by High-Code damage functions, while buildings of older
construction would be best represented by Moderate-Code damage functions, if built after about
1940, or by Pre-Code damage functions, if built before about 1940 (i.e., before seismic codes
existed). Pre-Code damage functions are appropriate for modeling older buildings that were not
designed for earthquake load, regardless of where they are located in the United States. Guidance
is provided to expert users in Section 5.7 for selection of appropriate building damage functions

5.4.2 Capacity Curves

Most buildings are presently designed or evaluated using linear-elastic analysis methods,
primarily due to the relative simplicity of these methods in comparison to more complex,
nonlinear methods. Typically, building response is based on linear-elastic properties of the
structure and forces corresponding to the design-basis earthquake. For design of building
elements, linear-elastic (5%-damped) response is reduced by a factor (e.g. the R-Factor in 1994
NEHRP Provisions) that varies for different types of lateral force resisting systems. The
reduction factor is based on empirical data and judgment that account for the inelastic
deformation capability (ductility) of the structural system, redundancy, overstrength, increased
damping (above 5% of critical) at large deformations, and other factors that influence building
capacity. Although this force-based approach is difficult to justify by rational engineering
analysis, buildings designed using these methods have performed reasonably well in past
earthquakes. Aspects of these methods found not to work well in earthquakes have been studied
and improved. In most cases, building capacity has been increased by improvements to detailing
practices (e.g., better confinement of steel reinforcement in concrete elements).

Except for a few brittle systems and acceleration-sensitive elements, building damage is primarily
a function of building displacement, rather than force. In the inelastic range of building response,
increasingly larger damage would result from increased building displacement although lateral
force would remain constant or decrease. Hence, successful prediction of earthquake damage to
buildings requires reasonably accurate estimation of building displacement response in the
inelastic range. This, however, can not be accomplished using linear-elastic methods, since the
buildings respond inelastically to earthquake ground shaking of magnitudes of interest for
damage prediction. Building capacity (push-over) curves, used with capacity spectrum method
(CSM) techniques [Mahaney, et. al., 1993, Kircher, 1996], provide simple and reasonably
accurate means of predicting inelastic building displacement response for damage estimation
purposes.

A building capacity curve (also known as a push-over curve) is a plot of a buildings lateral load
resistance as a function of a characteristic lateral displacement (i.e., a force-deflection plot). It is
1529
HazusMHTechnicalManual
derived from a plot of static-equivalent base shear versus building (e.g., roof) displacement. In
order to facilitate direct comparison with earthquake demand (i.e. overlaying the capacity curve
with a response spectrum), the force (base shear) axis is converted to spectral acceleration and the
displacement axis is converted to spectral displacement. Such a plot provides an estimate of the
buildings true deflection (displacement response) for any given earthquake response spectrum.

The building capacity curves developed for the Methodology are based on engineering design
parameters and judgment. Three control points that define model building capacity describe each
curve:

- DesignCapacity
- YieldCapacity
- UltimateCapacity

Design capacity represents the nominal building strength required by current model seismic code
provisions (e.g., 1994 NEHRP Provisions) or an estimate of the nominal strength for buildings
not designed for earthquake loads. Wind design is not considered in the estimation of design
capacity, and certain buildings (e.g., tall buildings located in zones of low or moderate seismicity)
may have a lateral design strength considerably greater than that based on seismic code
provisions.

Yield capacity represents the true lateral strength of the building considering redundancies in
design, conservatism in code requirements and true (rather than nominal) strength of materials.
Ultimate capacity represents the maximum strength of the building when the global structural
system has reached a fully plastic state. Ultimate capacity implicitly accounts for loss of strength
due to shear failure of brittle elements. Typically, buildings are assumed capable of deforming
beyond their ultimate point without loss of stability, but their structural system provides no
additional resistance to lateral earthquake force.

Up to the yield point, the building capacity curve is assumed to be linear with stiffness based on
an estimate of the true period of the building. The true period is typically longer than the code-
specified period of the building due to flexing of diaphragms of short, stiff buildings, flexural
cracking of elements of concrete and masonry structures, flexibility of foundations and other
factors observed to affect building stiffness. From the yield point to the ultimate point, the
capacity curve transitions in slope from an essentially elastic state to a fully plastic state. The
1530
Chapter5DirectPhysicalDamageGeneralBuildingStock
capacity curve is assumed to remain plastic past the ultimate point. An example building capacity
curve is shown in Figure 5.3.

Sa
Sd
Dy Du Dd
Au
Ay
Ad
Ultimate
Capacity
Yield
Capacity
Design
Capacity

Figure5.3ExampleBuildingCapacityCurve.

The building capacity curves are constructed based on estimates of engineering properties that
affect the design, yield and ultimate capacities of each model building type. These properties are
defined by the following parameters:

C
s
designstrengthcoefficient(fractionofbuildingsweight),
T
e
trueelasticfundamentalmodeperiodofbuilding(seconds),
o
1
fractionofbuildingweighteffectiveinpushovermode,
o
2
fractionofbuildingheightatlocationofpushovermodedisplacement,
overstrengthfactorrelatingtrueyieldstrengthtodesignstrength,
overstrengthfactorrelatingultimatestrengthtoyieldstrength,and
ductility factor relating ultimate displacement to times the yield
displacement(i.e.,assumedpointofsignificantyieldingofthestructure)
1531
HazusMHTechnicalManual

The design strength, C
s
, is approximately based, on the lateral-force design requirements of
current seismic codes (e.g., 1994 NEHRP Provisions). These requirements are a function of the
buildings seismic zone location and other factors including: site soil condition, type of lateral-
force-resisting system and building period. For each of the four design levels (High-Code,
Moderate-Code, Low-Code and Pre-Code), design capacity is based on the best estimate of
typical design properties. Table 5.4 summarizes design capacity for each building type and
design level. Building period, T
e
, push-over mode parameters o
1
and o
2
, the ratio of yield to
design strength, , and the ratio of ultimate to yield strength, , are assumed to be independent of
design level. Values of these parameters are summarized in Table 5.5 for each building type.
Values of the ductility factor, , are given in Table 5.6 for each building type and design level.
Note that for the following tables, shaded boxes indicate types that are not permitted by current
seismic codes.

1532
Chapter5DirectPhysicalDamageGeneralBuildingStock
Table5.4CodeBuildingCapacityParametersDesignStrength(C
s
)
Building SeismicDesignLevel(FractionofBuildingWeight)
Type
HighCode ModerateCode LowCode PreCode
W1 0.200 0.150 0.100 0.100
W2 0.200 0.100 0.050 0.050
S1L 0.133 0.067 0.033 0.033
S1M 0.100 0.050 0.025 0.025
S1H 0.067 0.033 0.017 0.017
S2L 0.200 0.100 0.050 0.050
S2M 0.200 0.100 0.050 0.050
S2H 0.150 0.075 0.038 0.038
S3 0.200 0.100 0.050 0.050
S4L 0.160 0.080 0.040 0.040
S4M 0.160 0.080 0.040 0.040
S4H 0.120 0.060 0.030 0.030
S5L 0.050 0.050
S5M 0.050 0.050
S5H 0.038 0.038
C1L 0.133 0.067 0.033 0.033
C1M 0.133 0.067 0.033 0.033
C1H 0.067 0.033 0.017 0.017
C2L 0.200 0.100 0.050 0.050
C2M 0.200 0.100 0.050 0.050
C2H 0.150 0.075 0.038 0.038
C3L 0.050 0.050
C3M 0.050 0.050
C3H 0.038 0.038
PC1 0.200 0.100 0.050 0.050
PC2L 0.200 0.100 0.050 0.050
PC2M 0.200 0.100 0.050 0.050
PC2H 0.150 0.075 0.038 0.038
RM1L 0.267 0.133 0.067 0.067
RM1M 0.267 0.133 0.067 0.067
RM2L 0.267 0.133 0.067 0.067
RM2M 0.267 0.133 0.067 0.067
RM2H 0.200 0.100 0.050 0.050
URML 0.067 0.067
URMM 0.067 0.067
1533
HazusMHTechnicalManual
Building SeismicDesignLevel(FractionofBuildingWeight)
Type
HighCode ModerateCode LowCode PreCode
MH 0.100 0.100 0.100 0.100



1534
Chapter5DirectPhysicalDamageGeneralBuildingStock
Table5.5CodeBuildingCapacityParametersPeriod(T
e
),PushoverMode
ResponseFactors(o
1
,o
2
)andOverstrengthRatios(,)
Building Heightto Period,T
e
ModalFactors OverstrengthRatios
Type
Roof(Feet) (Seconds) Weight,o
1
Height,o
2
Yield, Ultimate,
W1 14.0 0.35 0.75 0.75 1.50 3.00
W2 24.0 0.40 0.75 0.75 1.50 2.50
S1L 24.0 0.50 0.80 0.75 1.50 3.00
S1M 60.0 1.08 0.80 0.75 1.25 3.00
S1H 156.0 2.21 0.75 0.60 1.10 3.00
S2L 24.0 0.40 0.75 0.75 1.50 2.00
S2M 60.0 0.86 0.75 0.75 1.25 2.00
S2H 156.0 1.77 0.65 0.60 1.10 2.00
S3 15.0 0.40 0.75 0.75 1.50 2.00
S4L 24.0 0.35 0.75 0.75 1.50 2.25
S4M 60.0 0.65 0.75 0.75 1.25 2.25
S4H 156.0 1.32 0.65 0.60 1.10 2.25
S5L 24.0 0.35 0.75 0.75 1.50 2.00
S5M 60.0 0.65 0.75 0.75 1.25 2.00
S5H 156.0 1.32 0.65 0.60 1.10 2.00
C1L 20.0 0.40 0.80 0.75 1.50 3.00
C1M 50.0 0.75 0.80 0.75 1.25 3.00
C1H 120.0 1.45 0.75 0.60 1.10 3.00
C2L 20.0 0.35 0.75 0.75 1.50 2.50
C2M 50.0 0.56 0.75 0.75 1.25 2.50
C2H 120.0 1.09 0.65 0.60 1.10 2.50
C3L 20.0 0.35 0.75 0.75 1.50 2.25
C3M 50.0 0.56 0.75 0.75 1.25 2.25
C3H 120.0 1.09 0.65 0.60 1.10 2.25
PC1 15.0 0.35 0.50 0.75 1.50 2.00
PC2L 20.0 0.35 0.75 0.75 1.50 2.00
PC2M 50.0 0.56 0.75 0.75 1.25 2.00
PC2H 120.0 1.09 0.65 0.60 1.10 2.00
RM1L 20.0 0.35 0.75 0.75 1.50 2.00
RM1M 50.0 0.56 0.75 0.75 1.25 2.00
RM2L 20.0 0.35 0.75 0.75 1.50 2.00
RM2M 50.0 0.56 0.75 0.75 1.25 2.00
RM2H 120.0 1.09 0.65 0.60 1.10 2.00
1535
HazusMHTechnicalManual
Building Heightto Period,T
e
ModalFactors OverstrengthRatios
Type
Roof(Feet) (Seconds) Weight,o
1
Height,o
2
Yield, Ultimate,
URML 15.0 0.35 0.50 0.75 1.50 2.00
URMM 35.0 0.50 0.75 0.75 1.25 2.00
MH 10.0 0.35 1.00 1.00 1.50 2.00

1536
Chapter5DirectPhysicalDamageGeneralBuildingStock
Table5.6CodeBuildingCapacityParameterDuctility()
Building SeismicDesignLevel
Type
HighCode ModerateCode LowCode PreCode
W1 8.0 6.0 6.0 6.0
W2 8.0 6.0 6.0 6.0
S1L 8.0 6.0 5.0 5.0
S1M 5.3 4.0 3.3 3.3
S1H 4.0 3.0 2.5 2.5
S2L 8.0 6.0 5.0 5.0
S2M 5.3 4.0 3.3 3.3
S2H 4.0 3.0 2.5 2.5
S3 8.0 6.0 5.0 5.0
S4L 8.0 6.0 5.0 5.0
S4M 5.3 4.0 3.3 3.3
S4H 4.0 3.0 2.5 2.5
S5L 5.0 5.0
S5M 3.3 3.3
S5H 2.5 2.5
C1L 8.0 6.0 5.0 5.0
C1M 5.3 4.0 3.3 3.3
C1H 4.0 3.0 2.5 2.5
C2L 8.0 6.0 5.0 5.0
C2M 5.3 4.0 3.3 3.3
C2H 4.0 3.0 2.5 2.5
C3L 5.0 5.0
C3M 3.3 3.3
C3H 2.5 2.5
PC1 8.0 6.0 5.0 5.0
PC2L 8.0 6.0 5.0 5.0
PC2M 5.3 4.0 3.3 3.3
PC2H 4.0 3.0 2.5 2.5
RM1L 8.0 6.0 5.0 5.0
RM1M 5.3 4.0 3.3 3.3
RM2L 8.0 6.0 5.0 5.0
RM2M 5.3 4.0 3.3 3.3
RM2H 4.0 3.0 2.5 2.5
URML 5.0 5.0
URMM 3.3 3.3
1537
HazusMHTechnicalManual
Building SeismicDesignLevel
Type
HighCode ModerateCode LowCode PreCode
MH 6.0 6.0 6.0 6.0
1538
Chapter5DirectPhysicalDamageGeneralBuildingStock
Building capacity curves are assumed to have a range of possible properties that are lognormally
distributed as a function of the ultimate strength (A
u
) of each capacity curve. Capacity curves
described by the values of parameters given in Tables 5.4, 5.5 and 5.6 represent median estimates
of building capacity. The variability of the capacity of each building type is assumed to be: |(A
u
)
=0.25 for code-designed buildings (High-Code, Moderate-Code and Low-Code seismic design
levels) and |(A
u
) =0.30 for Pre-Code buildings.

Example construction of median, 84th percentile (+1|) and 16th percentile (-1|) building
capacity curves for a typical building is illustrated in Figure 5.4. Median capacity curves are
intersected with demand spectra to estimate peak building response. The variability of the
capacity curves is used, with other sources of variability and uncertainty, to define total fragility
curve variability.

Figure5.4ExampleConstructionofMedian,+1|and1|BuildingCapacityCurves.

Tables 5.7a, 5.7b, 5.7c and 5.7d summarize yield capacity and ultimate capacity control points for
High-Code, Moderate-Code, Low-Code and Pre-Code seismic design levels, respectively. Note
that for the following tables, shaded boxes indicate types that are not permitted by current seismic
codes.
1539
HazusMHTechnicalManual

1540
Chapter5DirectPhysicalDamageGeneralBuildingStock
Table5.7aCodeBuildingCapacityCurvesHighCodeSeismicDesignLevel
Building YieldCapacityPoint UltimateCapacityPoint
Type
D
y
(in.) A
y
(g) D
u
(in.) A
u
(g)
W1 0.48 0.400 11.51 1.200
W2 0.63 0.400 12.53 1.000
S1L 0.61 0.250 14.67 0.749
S1M 1.78 0.156 28.40 0.468
S1H 4.66 0.098 55.88 0.293
S2L 0.63 0.400 10.02 0.800
S2M 2.43 0.333 25.88 0.667
S2H 7.75 0.254 61.97 0.508
S3 0.63 0.400 10.02 0.800
S4L 0.38 0.320 6.91 0.720
S4M 1.09 0.267 13.10 0.600
S4H 3.49 0.203 31.37 0.457
S5L
S5M
S5H
C1L 0.39 0.250 9.39 0.749
C1M 1.15 0.208 18.44 0.624
C1H 2.01 0.098 24.13 0.293
C2L 0.48 0.400 9.59 1.000
C2M 1.04 0.333 13.84 0.833
C2H 2.94 0.254 29.39 0.635
C3L
C3M
C3H
PC1 0.72 0.600 11.51 1.200
PC2L 0.48 0.400 7.67 0.800
PC2M 1.04 0.333 11.07 0.667
PC2H 2.94 0.254 23.52 0.508
RM1L 0.64 0.533 10.23 1.066
RM1M 1.38 0.444 14.76 0.889
RM2L 0.64 0.533 10.23 1.066
RM2M 1.38 0.444 14.76 0.889
RM2H 3.92 0.338 31.35 0.677
URML
URMM
1541
HazusMHTechnicalManual
Building YieldCapacityPoint UltimateCapacityPoint
Type
D
y
(in.) A
y
(g) D
u
(in.) A
u
(g)
MH 0.18 0.150 2.16 0.300

1542
Chapter5DirectPhysicalDamageGeneralBuildingStock
Table5.7bCodeBuildingCapacityCurvesModerateCodeSeismicDesignLevel
Building YieldCapacityPoint UltimateCapacityPoint
Type
D
y
(in.) A
y
(g) D
u
(in.) A
u
(g)
W1 0.36 0.300 6.48 0.900
W2 0.31 0.200 4.70 0.500
S1L 0.31 0.125 5.50 0.375
S1M 0.89 0.078 10.65 0.234
S1H 2.33 0.049 20.96 0.147
S2L 0.31 0.200 3.76 0.400
S2M 1.21 0.167 9.70 0.333
S2H 3.87 0.127 23.24 0.254
S3 0.31 0.200 3.76 0.400
S4L 0.19 0.160 2.59 0.360
S4M 0.55 0.133 4.91 0.300
S4H 1.74 0.102 11.76 0.228
S5L
S5M
S5H
C1L 0.20 0.125 3.52 0.375
C1M 0.58 0.104 6.91 0.312
C1H 1.01 0.049 9.05 0.147
C2L 0.24 0.200 3.60 0.500
C2M 0.52 0.167 5.19 0.417
C2H 1.47 0.127 11.02 0.317
C3L
C3M
C3H
PC1 0.36 0.300 4.32 0.600
PC2L 0.24 0.200 2.88 0.400
PC2M 0.52 0.167 4.15 0.333
PC2H 1.47 0.127 8.82 0.254
RM1L 0.32 0.267 3.84 0.533
RM1M 0.69 0.222 5.54 0.444
RM2L 0.32 0.267 3.84 0.533
RM2M 0.69 0.222 5.54 0.444
RM2H 1.96 0.169 11.76 0.338
URML
URMM
1543
HazusMHTechnicalManual
Building YieldCapacityPoint UltimateCapacityPoint
Type
D
y
(in.) A
y
(g) D
u
(in.) A
u
(g)
MH 0.18 0.150 2.16 0.300

1544
Chapter5DirectPhysicalDamageGeneralBuildingStock
Table5.7cCodeBuildingCapacityCurvesLowCodeSeismicDesignLevel
Building YieldCapacityPoint UltimateCapacityPoint
Type
D
y
(in.) A
y
(g) D
u
(in.) A
u
(g)
W1 0.24 0.200 4.32 0.600
W2 0.16 0.100 2.35 0.250
S1L 0.15 0.062 2.29 0.187
S1M 0.44 0.039 4.44 0.117
S1H 1.16 0.024 8.73 0.073
S2L 0.16 0.100 1.57 0.200
S2M 0.61 0.083 4.04 0.167
S2H 1.94 0.063 9.68 0.127
S3 0.16 0.100 1.57 0.200
S4L 0.10 0.080 1.08 0.180
S4M 0.27 0.067 2.05 0.150
S4H 0.87 0.051 4.90 0.114
S5L 0.12 0.100 1.20 0.200
S5M 0.34 0.083 2.27 0.167
S5H 1.09 0.063 5.45 0.127
C1L 0.10 0.062 1.47 0.187
C1M 0.29 0.052 2.88 0.156
C1H 0.50 0.024 3.77 0.073
C2L 0.12 0.100 1.50 0.250
C2M 0.26 0.083 2.16 0.208
C2H 0.74 0.063 4.59 0.159
C3L 0.12 0.100 1.35 0.225
C3M 0.26 0.083 1.95 0.188
C3H 0.74 0.063 4.13 0.143
PC1 0.18 0.150 1.80 0.300
PC2L 0.12 0.100 1.20 0.200
PC2M 0.26 0.083 1.73 0.167
PC2H 0.74 0.063 3.67 0.127
RM1L 0.16 0.133 1.60 0.267
RM1M 0.35 0.111 2.31 0.222
RM2L 0.16 0.133 1.60 0.267
RM2M 0.35 0.111 2.31 0.222
RM2H 0.98 0.085 4.90 0.169
URML 0.24 0.200 2.40 0.400
URMM 0.27 0.111 1.81 0.222
1545
HazusMHTechnicalManual
Building YieldCapacityPoint UltimateCapacityPoint
Type
D
y
(in.) A
y
(g) D
u
(in.) A
u
(g)
MH 0.18 0.150 2.16 0.300

1546
Chapter5DirectPhysicalDamageGeneralBuildingStock
Table5.7dBuildingCapacityCurvesPreCodeSeismicDesignLevel
Building YieldCapacityPoint UltimateCapacityPoint
Type
D
y
(in.) A
y
(g) D
u
(in.) A
u
(g)
W1 0.24 0.200 4.32 0.600
W2 0.16 0.100 2.35 0.250
S1L 0.15 0.062 2.75 0.187
S1M 0.44 0.039 5.33 0.117
S1H 1.16 0.024 10.48 0.073
S2L 0.16 0.100 1.88 0.200
S2M 0.61 0.083 4.85 0.167
S2H 1.94 0.063 11.62 0.127
S3 0.16 0.100 1.88 0.200
S4L 0.10 0.080 1.30 0.180
S4M 0.27 0.067 2.46 0.150
S4H 0.87 0.051 5.88 0.114
S5L 0.12 0.100 1.20 0.200
S5M 0.34 0.083 2.27 0.167
S5H 1.09 0.063 5.45 0.127
C1L 0.10 0.062 1.76 0.187
C1M 0.29 0.052 3.46 0.156
C1H 0.50 0.024 4.52 0.073
C2L 0.12 0.100 1.80 0.250
C2M 0.26 0.083 2.60 0.208
C2H 0.74 0.063 5.51 0.159
C3L 0.12 0.100 1.35 0.225
C3M 0.26 0.083 1.95 0.188
C3H 0.74 0.063 4.13 0.143
PC1 0.18 0.150 2.16 0.300
PC2L 0.12 0.100 1.44 0.200
PC2M 0.26 0.083 2.08 0.167
PC2H 0.74 0.063 4.41 0.127
RM1L 0.16 0.133 1.92 0.267
RM1M 0.35 0.111 2.77 0.222
RM2L 0.16 0.133 1.92 0.267
RM2M 0.35 0.111 2.77 0.222
RM2H 0.98 0.085 5.88 0.169
URML 0.24 0.200 2.40 0.400
URMM 0.27 0.111 1.81 0.222
1547
HazusMHTechnicalManual
Building YieldCapacityPoint UltimateCapacityPoint
Type
D
y
(in.) A
y
(g) D
u
(in.) A
u
(g)
MH 0.18 0.150 2.16 0.300

5.4.3 Fragility Curves



This section describes building fragility curves for Slight, Moderate, Extensive and Complete
structural damage states and Slight, Moderate, Extensive and Complete nonstructural damage
states. Each fragility curve is characterized by median and lognormal standard deviation (|)
values of PESH demand. Spectral displacement is the PESH parameter used for structural
damage and nonstructural damage to drift-sensitive components. Spectral acceleration is the
PESH parameter used for calculating nonstructural damage to acceleration-sensitive components.

5.4.3.1 Background

The probability of being in or exceeding a given damage state is modeled as a cumulative
lognormal distribution. For structural damage, given the spectral displacement, S
d,
the probability
of being in or exceeding a damage state, ds, is modeled as:


| |
P S
1
ln
S
S
d
d
,
ds
ds
d ds
=
|
\

|
.
|

(
u
|
(5-3)

where: S
d,ds
isthemedianvalueofspectraldisplacementatwhichthe building
reachesthethresholdofthedamagestate,ds,
|
ds
is the standard deviation of the natural logarithm of spectral
displacementofdamagestate,ds,and
u isthestandardnormalcumulativedistributionfunction.

For example, a mid-rise, concrete-frame building (C1M) of High-Code seismic design has
Extensive structural damage defined by a median spectral displacement value (S
d,E
) of 9.0
1548
Chapter5DirectPhysicalDamageGeneralBuildingStock
inches and a lognormal standard deviation value (|
E
) of 0.68. The lognormal fragility curve for
Extensive structural damage to this building is shown in Figure 5.5.

In Figure 5.5, the symbol, S, indicates the median value of 9.0 inches. The symbol, S
+
, indicates
the +1 lognormal standard deviation level of the fragility curve, which is evaluated as S
+
=S x
exp(|) =17.8 inches. Similarly, the symbol, S
-
, indicates the -1 lognormal standard deviation
level of the fragility curve, which is evaluated as S
-
=S/exp(|) =4.6 inches. The corresponding
probabilities of being in or exceeding the Extensive damage state for this example are:

| |
| |
| |
P Extensive Damage
Extensive Damage S
Extensive Damage
= =
=
=

+
S S =4. inches 0.16
P S = . inches =0.50
P S S = . inches =0.84
d
d
d
6
90
178


Figure5.5ExampleFragilityCurveExtensiveStructuralDamage,
C1MModelBuildingType,HighCodeSeismicDesign.

1549
HazusMHTechnicalManual
5.4.3.2DevelopmentofDamageStateMedians

Median values of fragility curves are developed for each damage states (i.e., Slight, Moderate,
Extensive and Complete) and for each of the three types of building components: structural,
nonstructural drift-sensitive and nonstructural acceleration-sensitive components. Structural
fragility is characterized in terms of spectral displacement and by equivalent-PGA fragility curves
(for buildings that are components of lifelines). Section 5.4.4 describes development of median
values of equivalent-PGA structural fragility curves based on the structural fragility curves of this
section.

Median values of structural component fragility are based on building drift ratios that describe the
threshold of damage states. Damage-state drift ratios are converted to spectral displacement
using Equation (5-4):

S h
dSds R Sds , ,
= o o
2
(54)

where: S
dSds ,
isthemedianvalueofspectraldisplacement,ininches,of
structuralcomponentsfordamagestate,ds,
o
R,Sds
isthedriftratioatthethresholdofstructuraldamagestate,ds,
o
2
is the fraction of the building (roof) height at the location of pushover
modedisplacement,asspecifiedinTable5.5,and
h is the typical roof height, in inches, of the model building type of
interest(seeTable5.1fortypicalbuildingheight).

Values of damage-state drift ratios are included in the Methodology based, in part, on a study by
OAK Engineering [OAK, 1994] that reviewed and synthesized available drift/damage
information from a number of published sources, including Kustu et al. (1982), Ferritto (1982 and
1983), Czarnecki (1973), Hasselman et al. (1980), Whitman et al. (1977) and Wong (1975).

Median values of nonstructural drift-sensitive component fragility are based on building drift
ratios that describe the threshold of damage states. Nonstructural drift-sensitive components are
identified in Table 5.2. Damage state drift ratios for nonstructural drift-sensitive components are
1550
Chapter5DirectPhysicalDamageGeneralBuildingStock
converted to median values of spectral displacement using the same approach as that of Equation
(5-4). Values of damage-state drift are based, in part, on the work of Ferrito (1982 and 1983) and
on a recent update of this data included in a California Division of the State Architect report
[DSA, 1996].

Median values of nonstructural acceleration-sensitive component fragility are based on peak floor
(input) acceleration that describes the threshold of damage states. These values of acceleration
are used directly as median values of spectral acceleration for nonstructural acceleration-sensitive
component fragility curves. Values of damage-state acceleration are based, in part, on the work
of Ferrito (1982 and 1983) and on a recent update of this data included in a California Division of
the State Architect report [DSA, 1996].

5.4.3.3 Development of Damage State Variability

Lognormal standard deviation (|) values that describe the variability of fragility curves are
developed for each damage states (i.e., Slight, Moderate, Extensive and Complete) and for each
of the three types of building components: structural, nonstructural drift-sensitive and
nonstructural acceleration-sensitive components. Structural fragility is characterized in terms of
spectral displacement and by equivalent-PGA fragility curves (for buildings that are components
of lifelines). Section 5.4.4 describes development of variability values for equivalent-PGA
structural fragility curves.

The total variability of each structural damage state, |
Sds
, is modeled by the combination of three
contributors to structural damage variability, |
C
, |
D
and |
M(Sds)
, as described in Equation (5-5):

| | ( ) ( )
| | | |
Sds C D dSds M Sds
CONV S = + , ,
, ( )
2 2
(55)

where: |
Sds
isthelognormalstandarddeviationthatdescribesthetotal
variabilityforstructuraldamagestate,ds,
|
C
is the lognormal standard deviation parameter that describes
thevariabilityofthecapacitycurve,
1551
HazusMHTechnicalManual
|
D
is the lognormal standard deviation parameter that describes
thevariabilityofthedemandspectrum,
|
M(Sds)
is the lognormal standard deviation parameter that describes
the uncertainty in the estimate of the median value of the
thresholdofstructuraldamagestate,ds.

The variability of building response depends jointly on demand and capacity (since capacity
curves are nonlinear). The function CONV in Equation (5-5) implies a complex process of
convolving probability distributions of the demand spectrum and the capacity curve, respectively.
Demand spectra and capacity curves are described probabilistically by median properties and
variability parameters, |
D
and |
C
, respectively. Capacity curves are defined for each building
type, but the demand spectrum is based on the PESH input spectrum whose shape is a function of
source/site conditions. For development of building fragility curves, the demand spectrum shape
represented Moderate duration ground shaking of a large-magnitude WUS earthquake at a soil
site.

The convolution process produces a surface that describes the probability of each
demand/capacity intersection point when the median demand spectrum is scaled to intersect the
median capacity curve at a given amplitude of response. Discrete values of the probabilistic
surface are summed along a line anchored to the damage state median of interest (e.g., S
d,Sds
) to
estimate the probability of reaching or exceeding the median value given building response at the
intersection point. This process is repeated for other intersection points to form a cumulative
description of the probability of reaching (or exceeding) the damage state of interest. A
lognormal function is fit to this cumulative curve yielding an estimate of the lognormal standard
deviation of the combined effect of demand and capacity variability on building fragility.

The lognormal standard deviation parameter that describes the uncertainty in the estimate of the
median value of the threshold of structural damage state ds, |
M(Sds)
, is assumed to be independent
of capacity and demand, and is added by the square-root-sum-of-the-squares (SRSS) method to
the lognormal standard deviation parameter representing the combined effects of demand and
capacity variability.

In the development of the damage state variability for implementation with the USGS
probabilistic seismic hazard curves, the procedure was modified. The USGS explicitly
incorporated the ground motion uncertainty in their Project 97 seismic hazard curves. (See
Chapter 4) These hazard curves were the basis for the Hazus PESH data used in the
Methodologys probabilistic analysis procedure. To avoid overestimation of the damage state
1552
Chapter5DirectPhysicalDamageGeneralBuildingStock
variability due to this double counting of ground motion uncertainty, the convolution process was
modified and reanalyzed. Modified damage state variability parameters were developed for each
probabilistic return period (a total of 8 return periods) and used when the probabilistic analysis
option is selected. Due to large amount of modified parameters, their values are not reproduced in
this chapter. To review the modified parameters, the user can access them via the Hazus
software [Analysis-Damage Functions-Buildings].

The process, described above for structural components, is the same approach used to estimate
the lognormal standard deviation for nonstructural drift-sensitive components. Nonstructural
acceleration-sensitive components are treated in a similar manner to nonstructural drift-sensitive
components, except that cumulative descriptions of the probability of reaching (or exceeding) the
damage state of interest are developed in terms of spectral acceleration (rather than spectra
displacement). Also, nonstructural acceleration-sensitive components are divided into two sub-
populations: (1) components at or near ground level and (2) components at upper floors or on the
roof. PGA, rather than spectral acceleration, is a more appropriate PESH input for components at
or near ground level. Fragility curves for nonstructural acceleration-sensitive components assume
50% (low-rise), 33% (mid-rise) or 20% (high-rise) of nonstructural components are located at, or
near, the ground floor, and represent a weighted combination of the probability of damage to
components located at, or near, ground level and components located at upper-floor levels of the
building.

5.4.3.4 StructuralDamage

Structural damage fragility curves for buildings are described by median values of drift that
define the thresholds of Slight, Moderate, Extensive and Complete damage states. In general,
these estimates of drift are different for each model building type (including height) and seismic
design level. Table 5.8 summarizes the ranges of drift ratios used to define structural damage for
various low-rise building types designed to current High-Code seismic provisions. A complete
listing of damage-state drift ratios for all building types and heights are provided for each seismic
design level in Tables 5.9a, 5.9b, 5.9c and 5.9d, respectively.

Table5.8TypicalDriftRatiosUsedtoDefineMedianValuesofStructuralDamage

SeismicDesign BuildingType DriftRatioattheThresholdofStructuralDamage


Level (LowRise) Slight Moderate Extensive Complete
HighCode W1/W2
C1L,S2L
RM1L/RM2L,PC1/PC2L
0.004
0.005
0.004
0.012
0.010
0.008
0.040
0.030
0.024
0.100
0.080
0.070
1553
HazusMHTechnicalManual
ModerateCode

W1/W2
C1L,S2L
RM1L/RM2L,PC1/PC2L
0.004
0.005
0.004
0.010
0.009
0.007
0.031
0.023
0.019
0.075
0.060
0.053
LowCode W1/W2
C1L,S2L
RM1L/RM2L,PC1/PC2L URML,
C3L,S5L
0.004
0.005
0.004
0.003
0.010
0.008
0.006
0.006
0.031
0.020
0.016
0.015
0.075
0.050
0.044
0.035
PreCode

W1/W2
C1L,S2L
RM1L/RM2L,PC1/PC2L URML,
C3L,S5L
0.003
0.004
0.003
0.002
0.008
0.006
0.005
0.005
0.025
0.016
0.013
0.012
0.060
0.040
0.035
0.028

In general, values of the drift ratio that define Complete damage to Moderate-Code buildings are
assumed to be 75% of the drift ratio that define Complete damage to High-Code buildings, and
values of the drift ratio that define Complete damage to Low-Code buildings are assumed to be
63% of the drift ratios that define Complete damage to High-Code buildings. These assumptions
are based on the recognition that post-yield capacity is significantly less in buildings designed
with limited ductile detailing. Values of the drift ratio that define Slight damage were assumed to
be the same for High-Code, Moderate-Code and Low-Code buildings, since this damage state
typically does not exceed the buildings elastic capacity.

Values of drift ratios that define Moderate and Extensive damage to Moderate-Code and Low-
Code buildings are selected such that their distribution between Slight and Complete damage-
state drift ratios is in proportion to the distribution of damage-state drift ratios for High-Code
buildings.

Values of Pre-Code building drift ratios are based on the drift ratios for Low-Code buildings,
reduced slightly to account for inferior performance anticipated for these older buildings. For
each damage state, the drift ratio of a Pre-Code building is assumed to be 80% of the drift ratio of
the Low-Code building of the same building type.

Drift ratios are reduced for taller buildings assuming that the deflected shape will not affect
uniform distribution of drift over the buildings height. For all damage states, drift ratios for mid-
rise buildings are assumed to be 67% of those of low-rise buildings of the same type, and drift
ratios for high-rise buildings are assumed to be 50% of those of low-rise buildings of the same
type. Since mid-rise and high-rise buildings are much taller than low-rise buildings, median
values of spectral displacement (i.e., drift ratio times height of building at the point of push-over
mode displacement) are still much greater for mid-rise and high-rise buildings than for low-rise
buildings.
1554
Chapter5DirectPhysicalDamageGeneralBuildingStock

The total variability of each structural damage state, |
Sds
, is modeled by the combination of
following three contributors to damage variability:
- uncertaintyinthedamagestatethresholdofthestructuralsystem(|
M(Sds)
=
0.4,forallstructuraldamagestatesandbuildingtypes)
- variability in capacity (response) properties of the model building
type/seismicdesignlevelofinterest(|
C(Au)
=0.25forCodebuildings,|
C(Au)
=
0.30forPreCodebuildings)and
- variabilityinresponseduetothespatialvariabilityofgroundmotion
Each of these three contributors to damage state variability is assumed to be lognormally
distributed random variables. Capacity and demand are dependent parameters and a convolution
process is used to derive combined capacity/demand variability of each structural damage state.
Capacity/demand variability is then combined with damage state uncertainty, as described in
Section 5.4.3.3.

Tables 5.9a, 5.9b, 5.9c and 5.9d summarize median and lognormal standard deviation (|
Sds
)
values for Slight, Moderate, Extensive and Complete structural damage states High-Code,
Moderate-Code, Low-Code and Pre-Code buildings, respectively. Note that for the following
tables, shaded boxes indicate types that are not permitted by current seismic codes.

1555
HazusMHTechnicalManual
Table5.9aStructuralFragilityCurveParametersHighCodeSeismicDesign
Level

BuildingProperties Interstory Drift at Spectral Displacement (inches)
Type Height (inches) Threshold of DamageState Slight Moderate Extensive Complete
Roof Modal Slight Moderate Extensive Complete Median Beta Median Beta Median Beta Median Beta
W1 168 126 0.0040 0.0120 0.0400 0.1000 0.50 0.80 1.51 0.81 5.04 0.85 12.60 0.97
W2 288 216 0.0040 0.0120 0.0400 0.1000 0.86 0.81 2.59 0.88 8.64 0.90 21.60 0.83
S1L 288 216 0.0060 0.0120 0.0300 0.0800 1.30 0.80 2.59 0.76 6.48 0.69 17.28 0.72
S1M 720 540 0.0040 0.0080 0.0200 0.0533 2.16 0.65 4.32 0.66 10.80 0.67 28.80 0.74
S1H 1872 1123 0.0030 0.0060 0.0150 0.0400 3.37 0.64 6.74 0.64 16.85 0.65 44.93 0.67
S2L 288 216 0.0050 0.0100 0.0300 0.0800 1.08 0.81 2.16 0.89 6.48 0.94 17.28 0.83
S2M 720 540 0.0033 0.0067 0.0200 0.0533 1.80 0.67 3.60 0.67 10.80 0.68 28.80 0.79
S2H 1872 1123 0.0025 0.0050 0.0150 0.0400 2.81 0.63 5.62 0.63 16.85 0.64 44.93 0.71
S3 180 135 0.0040 0.0080 0.0240 0.0700 0.54 0.81 1.08 0.82 3.24 0.91 9.45 0.90
S4L 288 216 0.0040 0.0080 0.0240 0.0700 0.86 0.89 1.73 0.89 5.18 0.98 15.12 0.87
S4M 720 540 0.0027 0.0053 0.0160 0.0467 1.44 0.77 2.88 0.72 8.64 0.70 25.20 0.89
S4H 1872 1123 0.0020 0.0040 0.0120 0.0350 2.25 0.64 4.49 0.66 13.48 0.69 39.31 0.77
S5L
S5M
S5H
C1L 240 180 0.0050 0.0100 0.0300 0.0800 0.90 0.81 1.80 0.84 5.40 0.86 14.40 0.81
C1M 600 450 0.0033 0.0067 0.0200 0.0533 1.50 0.68 3.00 0.67 9.00 0.68 24.00 0.81
C1H 1440 864 0.0025 0.0050 0.0150 0.0400 2.16 0.66 4.32 0.64 12.96 0.67 34.56 0.78
C2L 240 180 0.0040 0.0100 0.0300 0.0800 0.72 0.81 1.80 0.84 5.40 0.93 14.40 0.92
C2M 600 450 0.0027 0.0067 0.0200 0.0533 1.20 0.74 3.00 0.77 9.00 0.68 24.00 0.77
C2H 1440 864 0.0020 0.0050 0.0150 0.0400 1.73 0.68 4.32 0.65 12.96 0.66 34.56 0.75
C3L
C3M
C3H
PC1 180 135 0.0040 0.0080 0.0240 0.0700 0.54 0.76 1.08 0.86 3.24 0.88 9.45 0.99
PC2L 240 180 0.0040 0.0080 0.0240 0.0700 0.72 0.84 1.44 0.88 4.32 0.98 12.60 0.94
PC2M 600 450 0.0027 0.0053 0.0160 0.0467 1.20 0.77 2.40 0.81 7.20 0.70 21.00 0.82
PC2H 1440 864 0.0020 0.0040 0.0120 0.0350 1.73 0.64 3.46 0.66 10.37 0.68 30.24 0.81
RM1L 240 180 0.0040 0.0080 0.0240 0.0700 0.72 0.84 1.44 0.86 4.32 0.92 12.60 1.01
RM1M 600 450 0.0027 0.0053 0.0160 0.0467 1.20 0.71 2.40 0.81 7.20 0.76 21.00 0.75
RM2L 240 180 0.0040 0.0080 0.0240 0.0700 0.72 0.80 1.44 0.81 4.32 0.91 12.60 0.98
RM2M 600 450 0.0027 0.0053 0.0160 0.0467 1.20 0.71 2.40 0.79 7.20 0.70 21.00 0.73
RM2H 1440 864 0.0020 0.0040 0.0120 0.0350 1.73 0.66 3.46 0.65 10.37 0.66 30.24 0.72
URML
URMM
MH 120 120 0.0040 0.0080 0.0240 0.0700 0.48 0.91 0.96 1.00 2.88 1.03 8.40 0.92

1556
Chapter5DirectPhysicalDamageGeneralBuildingStock
Table5.9bStructuralFragilityCurveParametersModerateCodeSeismic
DesignLevel

BuildingProperties Interstory Drift at Spectral Displacement (inches)


Type Height (inches) Threshold of DamageState Slight Moderate Extensive Complete
Roof Modal Slight Moderate Extensive Complete Median Beta Median Beta Median Beta Median Beta
W1 168 126 0.0040 0.0099 0.0306 0.0750 0.50 0.84 1.25 0.86 3.86 0.89 9.45 1.04
W2 288 216 0.0040 0.0099 0.0306 0.0750 0.86 0.89 2.14 0.95 6.62 0.95 16.20 0.92
S1L 288 216 0.0060 0.0104 0.0235 0.0600 1.30 0.80 2.24 0.75 5.08 0.74 12.96 0.88
S1M 720 540 0.0040 0.0069 0.0157 0.0400 2.16 0.65 3.74 0.68 8.46 0.69 21.60 0.87
S1H 1872 1123 0.0030 0.0052 0.0118 0.0300 3.37 0.64 5.83 0.64 13.21 0.71 33.70 0.83
S2L 288 216 0.0050 0.0087 0.0233 0.0600 1.08 0.93 1.87 0.92 5.04 0.93 12.96 0.93
S2M 720 540 0.0033 0.0058 0.0156 0.0400 1.80 0.70 3.12 0.69 8.40 0.69 21.60 0.89
S2H 1872 1123 0.0025 0.0043 0.0117 0.0300 2.81 0.66 4.87 0.64 13.10 0.69 33.70 0.80
S3 180 135 0.0040 0.0070 0.0187 0.0525 0.54 0.88 0.94 0.92 2.52 0.97 7.09 0.89
S4L 288 216 0.0040 0.0069 0.0187 0.0525 0.86 0.96 1.50 1.00 4.04 1.03 11.34 0.92
S4M 720 540 0.0027 0.0046 0.0125 0.0350 1.44 0.75 2.50 0.72 6.73 0.72 18.90 0.94
S4H 1872 1123 0.0020 0.0035 0.0093 0.0262 2.25 0.66 3.90 0.67 10.50 0.70 29.48 0.90
S5L
S5M
S5H
C1L 240 180 0.0050 0.0087 0.0233 0.0600 0.90 0.89 1.56 0.90 4.20 0.90 10.80 0.89
C1M 600 450 0.0033 0.0058 0.0156 0.0400 1.50 0.70 2.60 0.70 7.00 0.70 18.00 0.89
C1H 1440 864 0.0025 0.0043 0.0117 0.0300 2.16 0.66 3.74 0.66 10.08 0.76 25.92 0.91
C2L 240 180 0.0040 0.0084 0.0232 0.0600 0.72 0.91 1.52 0.97 4.17 1.03 10.80 0.87
C2M 600 450 0.0027 0.0056 0.0154 0.0400 1.20 0.81 2.53 0.77 6.95 0.73 18.00 0.91
C2H 1440 864 0.0020 0.0042 0.0116 0.0300 1.73 0.66 3.64 0.68 10.00 0.70 25.92 0.87
C3L
C3M
C3H
PC1 180 135 0.0040 0.0070 0.0187 0.0525 0.54 0.89 0.94 0.92 2.52 0.97 7.09 1.04
PC2L 240 180 0.0040 0.0069 0.0187 0.0525 0.72 0.96 1.25 1.00 3.37 1.03 9.45 0.88
PC2M 600 450 0.0027 0.0046 0.0125 0.0350 1.20 0.82 2.08 0.79 5.61 0.75 15.75 0.93
PC2H 1440 864 0.0020 0.0035 0.0094 0.0263 1.73 0.68 3.00 0.69 8.08 0.77 22.68 0.89
RM1L 240 180 0.0040 0.0069 0.0187 0.0525 0.72 0.96 1.25 0.99 3.37 1.05 9.45 0.94
RM1M 600 450 0.0027 0.0046 0.0125 0.0350 1.20 0.81 2.08 0.82 5.61 0.80 15.75 0.89
RM2L 240 180 0.0040 0.0069 0.0187 0.0525 0.72 0.91 1.25 0.96 3.37 1.02 9.45 0.93
RM2M 600 450 0.0027 0.0046 0.0125 0.0350 1.20 0.81 2.08 0.80 5.61 0.75 15.75 0.88
RM2H 1440 864 0.0020 0.0035 0.0094 0.0263 1.73 0.67 3.00 0.69 8.08 0.70 22.68 0.86
URML
URMM
MH 120 120 0.0040 0.0080 0.0240 0.0700 0.48 0.91 0.96 1.00 2.88 1.03 8.40 0.92

1557
HazusMHTechnicalManual
Table5.9cStructuralFragilityCurveParametersLowCodeSeismicDesignLevel
BuildingProperties Interstory Drift at Spectral Displacement (inches)
Type Height (inches) Threshold of DamageState Slight Moderate Extensive Complete
Roof Modal Slight Moderate Extensive Complete Median Beta Median Beta Median Beta Median Beta
W1 168 126 0.0040 0.0099 0.0306 0.0750 0.50 0.93 1.25 0.98 3.86 1.02 9.45 0.99
W2 288 216 0.0040 0.0099 0.0306 0.0750 0.86 0.97 2.14 0.90 6.62 0.89 16.20 0.99
S1L 288 216 0.0060 0.0096 0.0203 0.0500 1.30 0.77 2.07 0.78 4.38 0.78 10.80 0.96
S1M 720 540 0.0040 0.0064 0.0135 0.0333 2.16 0.68 3.44 0.78 7.30 0.85 18.00 0.98
S1H 1872 1123 0.0030 0.0048 0.0101 0.0250 3.37 0.66 5.37 0.70 11.38 0.76 28.08 0.92
S2L 288 216 0.0050 0.0080 0.0200 0.0500 1.08 0.96 1.73 0.89 4.32 0.86 10.80 0.98
S2M 720 540 0.0033 0.0053 0.0133 0.0333 1.80 0.70 2.88 0.73 7.20 0.85 18.00 0.98
S2H 1872 1123 0.0025 0.0040 0.0100 0.0250 2.81 0.66 4.49 0.67 11.23 0.74 28.08 0.92
S3 180 135 0.0040 0.0064 0.0161 0.0438 0.54 0.98 0.87 0.99 2.17 1.01 5.91 0.90
S4L 288 216 0.0040 0.0064 0.0161 0.0438 0.86 1.05 1.38 0.98 3.47 0.89 9.45 0.98
S4M 720 540 0.0027 0.0043 0.0107 0.0292 1.44 0.76 2.31 0.78 5.78 0.90 15.75 0.99
S4H 1872 1123 0.0020 0.0032 0.0080 0.0219 2.25 0.70 3.60 0.75 9.01 0.90 24.57 0.98
S5L 288 216 0.0030 0.0060 0.0150 0.0350 0.65 1.11 1.30 1.04 3.24 0.99 7.56 0.95
S5M 720 540 0.0020 0.0040 0.0100 0.0233 1.08 0.77 2.16 0.79 5.40 0.87 12.60 0.98
S5H 1872 1123 0.0015 0.0030 0.0075 0.0175 1.68 0.70 3.37 0.73 8.42 0.89 19.66 0.97
C1L 240 180 0.0050 0.0080 0.0200 0.0500 0.90 0.95 1.44 0.91 3.60 0.85 9.00 0.97
C1M 600 450 0.0033 0.0053 0.0133 0.0333 1.50 0.70 2.40 0.74 6.00 0.86 15.00 0.98
C1H 1440 864 0.0025 0.0040 0.0100 0.0250 2.16 0.70 3.46 0.81 8.64 0.89 21.60 0.98
C2L 240 180 0.0040 0.0076 0.0197 0.0500 0.72 1.04 1.37 1.02 3.55 0.99 9.00 0.95
C2M 600 450 0.0027 0.0051 0.0132 0.0333 1.20 0.82 2.29 0.81 5.92 0.81 15.00 0.99
C2H 1440 864 0.0020 0.0038 0.0099 0.0250 1.73 0.68 3.30 0.73 8.53 0.84 21.60 0.95
C3L 240 180 0.0030 0.0060 0.0150 0.0350 0.54 1.09 1.08 1.07 2.70 1.08 6.30 0.91
C3M 600 450 0.0020 0.0040 0.0100 0.0233 0.90 0.85 1.80 0.83 4.50 0.79 10.50 0.98
C3H 1440 864 0.0015 0.0030 0.0075 0.0175 1.30 0.71 2.59 0.74 6.48 0.90 15.12 0.97
PC1 180 135 0.0040 0.0064 0.0161 0.0438 0.54 1.00 0.87 1.05 2.17 1.12 5.91 0.89
PC2L 240 180 0.0040 0.0064 0.0161 0.0438 0.72 1.08 1.15 1.03 2.89 0.98 7.88 0.96
PC2M 600 450 0.0027 0.0043 0.0107 0.0292 1.20 0.81 1.92 0.79 4.81 0.84 13.12 0.99
PC2H 1440 864 0.0020 0.0032 0.0080 0.0219 1.73 0.71 2.77 0.75 6.93 0.89 18.90 0.98
RM1L 240 180 0.0040 0.0064 0.0161 0.0438 0.72 1.11 1.15 1.10 2.89 1.10 7.88 0.92
RM1M 600 450 0.0027 0.0043 0.0107 0.0292 1.20 0.87 1.92 0.84 4.81 0.79 13.12 0.96
RM2L 240 180 0.0040 0.0064 0.0161 0.0438 0.72 1.05 1.15 1.07 2.89 1.09 7.88 0.91
RM2M 600 450 0.0027 0.0043 0.0107 0.0292 1.20 0.84 1.92 0.81 4.81 0.77 13.12 0.96
RM2H 1440 864 0.0020 0.0032 0.0080 0.0219 1.73 0.69 2.77 0.72 6.93 0.87 18.90 0.96
URML 180 135 0.0030 0.0060 0.0150 0.0350 0.41 0.99 0.81 1.05 2.03 1.10 4.73 1.08
URMM 420 315 0.0020 0.0040 0.0100 0.0233 0.63 0.91 1.26 0.92 3.15 0.87 7.35 0.91
MH 120 120 0.0040 0.0080 0.0240 0.0700 0.48 0.91 0.96 1.00 2.88 1.03 8.40 0.92

1558
Chapter5DirectPhysicalDamageGeneralBuildingStock
Table5.9dStructuralFragilityCurveParametersPreCodeSeismicDesignLevel

5.4.3.5 NonstructuralDamageDriftSensitiveComponents

Table 5.10 summarizes drift ratios used by the Methodology to define the median values of
damage fragility curves for drift-sensitive nonstructural components of buildings. Nonstructural
damage drift ratios are assumed to be the same for each building type and each seismic design
level.

Table5.10DriftRatiosUsedtoDefineMedianValuesofDamagefor
NonstructuralDriftSensitiveComponents
DriftRatioattheThresholdofNonstructuralDamage
Slight Moderate Extensive Complete
0.004 0.008 0.025 0.050

BuildingProperties Interstory Drift at Spectral Displacement (inches)


Type Height (inches) Thresholdof DamageState Slight Moderate Extensive Complete
Roof Modal Slight Moderate Extensive Complete Median Beta Median Beta Median Beta Median Beta
W1 168 126 0.0032 0.0079 0.0245 0.0600 0.40 1.01 1.00 1.05 3.09 1.07 7.56 1.06
W2 288 216 0.0032 0.0079 0.0245 0.0600 0.69 1.04 1.71 0.97 5.29 0.90 12.96 0.99
S1L 288 216 0.0048 0.0076 0.0162 0.0400 1.04 0.85 1.65 0.82 3.50 0.80 8.64 0.95
S1M 720 540 0.0032 0.0051 0.0108 0.0267 1.73 0.70 2.76 0.75 5.84 0.81 14.40 0.98
S1H 1872 1123 0.0024 0.0038 0.0081 0.0200 2.70 0.69 4.30 0.71 9.11 0.85 22.46 0.93
S2L 288 216 0.0040 0.0064 0.0160 0.0400 0.86 1.01 1.38 0.96 3.46 0.88 8.64 0.98
S2M 720 540 0.0027 0.0043 0.0107 0.0267 1.44 0.73 2.30 0.75 5.76 0.80 14.40 0.98
S2H 1872 1123 0.0020 0.0032 0.0080 0.0200 2.25 0.70 3.59 0.70 8.99 0.84 22.46 0.91
S3 180 135 0.0032 0.0051 0.0128 0.0350 0.43 1.06 0.69 1.03 1.73 1.07 4.73 0.89
S4L 288 216 0.0032 0.0051 0.0128 0.0350 0.69 1.11 1.11 1.03 2.77 0.99 7.56 0.98
S4M 720 540 0.0021 0.0034 0.0086 0.0233 1.15 0.81 1.85 0.80 4.62 0.94 12.60 1.00
S4H 1872 1123 0.0016 0.0026 0.0064 0.0175 1.80 0.73 2.88 0.75 7.21 0.90 19.66 0.97
S5L 288 216 0.0024 0.0048 0.0120 0.0280 0.52 1.20 1.04 1.11 2.59 1.08 6.05 0.95
S5M 720 540 0.0016 0.0032 0.0080 0.0187 0.86 0.85 1.73 0.83 4.32 0.94 10.08 0.99
S5H 1872 1123 0.0012 0.0024 0.0060 0.0140 1.35 0.72 2.70 0.75 6.74 0.92 15.72 0.96
C1L 240 180 0.0040 0.0064 0.0160 0.0400 0.72 0.98 1.15 0.94 2.88 0.90 7.20 0.97
C1M 600 450 0.0027 0.0043 0.0107 0.0267 1.20 0.73 1.92 0.77 4.80 0.83 12.00 0.98
C1H 1440 864 0.0020 0.0032 0.0080 0.0200 1.73 0.71 2.76 0.80 6.91 0.94 17.28 1.01
C2L 240 180 0.0032 0.0061 0.0158 0.0400 0.58 1.11 1.10 1.09 2.84 1.07 7.20 0.93
C2M 600 450 0.0021 0.0041 0.0105 0.0267 0.96 0.86 1.83 0.83 4.74 0.80 12.00 0.98
C2H 1440 864 0.0016 0.0031 0.0079 0.0200 1.38 0.73 2.64 0.75 6.82 0.92 17.28 0.97
C3L 240 180 0.0024 0.0048 0.0120 0.0280 0.43 1.19 0.86 1.15 2.16 1.15 5.04 0.92
C3M 600 450 0.0016 0.0032 0.0080 0.0187 0.72 0.90 1.44 0.86 3.60 0.90 8.40 0.96
C3H 1440 864 0.0012 0.0024 0.0060 0.0140 1.04 0.73 2.07 0.75 5.18 0.90 12.10 0.95
PC1 180 135 0.0032 0.0051 0.0128 0.0350 0.43 1.14 0.69 1.14 1.73 1.17 4.73 0.98
PC2L 240 180 0.0032 0.0051 0.0128 0.0350 0.58 1.14 0.92 1.10 2.31 1.10 6.30 0.93
PC2M 600 450 0.0021 0.0034 0.0086 0.0233 0.96 0.87 1.54 0.83 3.85 0.91 10.50 1.00
PC2H 1440 864 0.0016 0.0026 0.0064 0.0175 1.38 0.74 2.21 0.75 5.55 0.91 15.12 0.96
RM1L 240 180 0.0032 0.0051 0.0128 0.0350 0.58 1.20 0.92 1.17 2.31 1.17 6.30 0.94
RM1M 600 450 0.0021 0.0034 0.0086 0.0233 0.96 0.91 1.54 0.89 3.85 0.89 10.50 0.96
RM2L 240 180 0.0032 0.0051 0.0128 0.0350 0.58 1.14 0.92 1.10 2.31 1.15 6.30 0.92
RM2M 600 450 0.0021 0.0034 0.0086 0.0233 0.96 0.89 1.54 0.87 3.85 0.87 10.50 0.96
RM2H 1440 864 0.0016 0.0026 0.0064 0.0175 1.38 0.75 2.21 0.75 5.55 0.84 15.12 0.94
URML 180 135 0.0024 0.0048 0.0120 0.0280 0.32 1.15 0.65 1.19 1.62 1.20 3.78 1.18
URMM 420 315 0.0016 0.0032 0.0080 0.0187 0.50 0.99 1.01 0.97 2.52 0.90 5.88 0.88
MH 120 120 0.0032 0.0064 0.0192 0.0560 0.38 1.11 0.77 1.10 2.30 0.95 6.72 0.97

1559
HazusMHTechnicalManual
Median values of drift-sensitive nonstructural fragility curves are based on global building
displacement (in inches), calculated as the product of: (1) drift ratio, (2) building height and (3)
the fraction of building height at the location of push-over mode displacement (o
2
).

The total variability of each nonstructural drift-sensitive damage state, |
NSDds
, is modeled by the
combination of following three contributors to damage variability:

- uncertainty in the damagestate threshold of nonstructural components
(|
M(NSDds)
=0.5,foralldamagestatesandbuildingtypes),
- variability in capacity (response) properties of the model building type that
contains the nonstructural components of interest (|
C(Au)
= 0.25 for Code
buildings,|
C(Au)
=0.30forPreCodebuildings),and
- variability in response of the model building type due to the spatial
variabilityofgroundmotiondemand(|
D(A)
=0.45and|
C(V)
=0.50).

Each of these three contributors to damage state variability is assumed to be lognormally


distributed random variables. Capacity and demand are dependent parameters and a convolution
process is used to derive combined capacity/demand variability of each nonstructural damage
state. Capacity/demand variability is then combined with damage state uncertainty, as described
in Section 5.4.3.3.

Table 5.11a, 5.11b, 5.11c and 5.11d summarize median and lognormal standard deviation (|
NSDds
)
values for Slight, Moderate, Extensive and Complete nonstructural drift-sensitive damage states
for High-Code, Moderate-Code, Low-Code and Pre-Code buildings, respectively. Median values
are the same for all design levels. Lognormal standard deviation values are slightly different for
each seismic design level. Note that for the following tables, shaded boxes indicate types that are
not permitted by current seismic codes.

1560
Chapter5DirectPhysicalDamageGeneralBuildingStock
Table5.11NonstructuralDriftSensitiveFragilityCurveParameters
HighCodeSeismicDesignLevel
Building MedianSpectralDisplacement(inches)andLogstandardDeviation(Beta)
Type Slight Moderate Extensive Complete
Median Beta Median Beta Median Beta Median Beta
W1 0.50 0.85 1.01 0.88 3.15 0.88 6.30 0.94
W2 0.86 0.87 1.73 0.89 5.40 0.96 10.80 0.94
S1L 0.86 0.81 1.73 0.85 5.40 0.77 10.80 0.77
S1M 2.16 0.71 4.32 0.72 13.50 0.72 27.00 0.80
S1H 4.49 0.72 8.99 0.71 28.08 0.74 56.16 0.77
S2L 0.86 0.84 1.73 0.90 5.40 0.97 10.80 0.92
S2M 2.16 0.71 4.32 0.74 13.50 0.74 27.00 0.84
S2H 4.49 0.71 8.99 0.71 28.08 0.72 56.16 0.78
S3 0.54 0.86 1.08 0.88 3.38 0.98 6.75 0.98
S4L 0.86 0.93 1.73 0.94 5.40 1.01 10.80 0.99
S4M 2.16 0.80 4.32 0.76 13.50 0.76 27.00 0.93
S4H 4.49 0.72 8.99 0.72 28.08 0.79 56.16 0.91
S5L
S5M
S5H
C1L 0.72 0.84 1.44 0.88 4.50 0.90 9.00 0.88
C1M 1.80 0.72 3.60 0.73 11.25 0.74 22.50 0.84
C1H 3.46 0.71 6.91 0.71 21.60 0.78 43.20 0.88
C2L 0.72 0.87 1.44 0.88 4.50 0.97 9.00 0.99
C2M 1.80 0.84 3.60 0.82 11.25 0.74 22.50 0.81
C2H 3.46 0.71 6.91 0.72 21.60 0.74 43.20 0.85
C3L
C3M
C3H
PC1 0.54 0.82 1.08 0.91 3.38 0.95 6.75 1.03
PC2L 0.72 0.89 1.44 0.93 4.50 1.03 9.00 1.04
PC2M 1.80 0.87 3.60 0.83 11.25 0.77 22.50 0.89
PC2H 3.46 0.73 6.91 0.73 21.60 0.77 43.20 0.89
RM1L 0.72 0.89 1.44 0.91 4.50 0.97 9.00 1.06
RM1M 1.80 0.81 3.60 0.86 11.25 0.80 22.50 0.81
RM2L 0.72 0.85 1.44 0.87 4.50 0.95 9.00 1.03
RM2M 1.80 0.82 3.60 0.84 11.25 0.76 22.50 0.80
1561
HazusMHTechnicalManual
RM2H 3.46 0.71 6.91 0.73 21.60 0.73 43.20 0.85
URML
URMM
MH 0.48 0.96 0.96 1.05 3.00 1.07 6.00 0.93
1562
Chapter5DirectPhysicalDamageGeneralBuildingStock
Table5.11bNonstructuralDriftSensitiveFragilityCurveParameters
ModerateCodeSeismicDesignLevel
Building MedianSpectralDisplacement(inches)andLogstandardDeviation(Beta)
Type Slight Moderate Extensive Complete
Median Beta Median Beta Median Beta Median Beta
W1 0.50 0.89 1.01 0.91 3.15 0.90 6.30 1.04
W2 0.86 0.94 1.73 0.99 5.40 1.00 10.80 0.90
S1L 0.86 0.84 1.73 0.83 5.40 0.79 10.80 0.87
S1M 2.16 0.71 4.32 0.74 13.50 0.85 27.00 0.95
S1H 4.49 0.71 8.99 0.74 28.08 0.84 56.16 0.95
S2L 0.86 0.93 1.73 0.99 5.40 0.96 10.80 0.92
S2M 2.16 0.74 4.32 0.74 13.50 0.85 27.00 0.96
S2H 4.49 0.72 8.99 0.73 28.08 0.80 56.16 0.94
S3 0.54 0.93 1.08 0.98 3.38 1.01 6.75 0.94
S4L 0.86 1.00 1.73 1.06 5.40 0.99 10.80 0.96
S4M 2.16 0.77 4.32 0.80 13.50 0.95 27.00 1.04
S4H 4.49 0.73 8.99 0.82 28.08 0.93 56.16 1.01
S5L
S5M
S5H
C1L 0.72 0.93 1.44 0.96 4.50 0.94 9.00 0.88
C1M 1.80 0.77 3.60 0.76 11.25 0.87 22.50 0.98
C1H 3.46 0.74 6.91 0.80 21.60 0.94 43.20 1.03
C2L 0.72 0.96 1.44 1.00 4.50 1.06 9.00 0.95
C2M 1.80 0.84 3.60 0.81 11.25 0.83 22.50 0.98
C2H 3.46 0.73 6.91 0.76 21.60 0.89 43.20 0.99
C3L
C3M
C3H
PC1 0.54 0.94 1.08 0.99 3.38 1.05 6.75 1.08
PC2L 0.72 1.00 1.44 1.06 4.50 1.07 9.00 0.93
PC2M 1.80 0.85 3.60 0.83 11.25 0.92 22.50 1.00
PC2H 3.46 0.74 6.91 0.79 21.60 0.93 43.20 1.02
RM1L 0.72 1.00 1.44 1.06 4.50 1.12 9.00 1.01
RM1M 1.80 0.88 3.60 0.85 11.25 0.84 22.50 0.98
RM2L 0.72 0.96 1.44 1.02 4.50 1.10 9.00 0.99
RM2M 1.80 0.88 3.60 0.83 11.25 0.81 22.50 0.98
1563
HazusMHTechnicalManual
RM2H 3.46 0.73 6.91 0.76 21.60 0.88 43.20 0.99
URML
URMM
MH 0.48 0.96 0.96 1.05 3.00 1.07 6.00 0.93
1564
Chapter5DirectPhysicalDamageGeneralBuildingStock
Table5.11cNonstructuralDriftSensitiveFragilityCurveParameters
LowCodeSeismicDesignLevel
Building MedianSpectralDisplacement(inches)andLogstandardDeviation(Beta)
Type Slight Moderate Extensive Complete
Median Beta Median Beta Median Beta Median Beta
W1 0.50 0.98 1.01 0.99 3.15 1.02 6.30 1.09
W2 0.86 1.01 1.73 0.97 5.40 0.93 10.80 1.03
S1L 0.86 0.86 1.73 0.84 5.40 0.88 10.80 1.00
S1M 2.16 0.74 4.32 0.89 13.50 0.99 27.00 1.05
S1H 4.49 0.75 8.99 0.87 28.08 0.97 56.16 1.04
S2L 0.86 1.01 1.73 0.94 5.40 0.94 10.80 1.03
S2M 2.16 0.77 4.32 0.87 13.50 0.99 27.00 1.05
S2H 4.49 0.74 8.99 0.86 28.08 0.97 56.16 1.05
S3 0.54 1.03 1.08 1.02 3.38 0.96 6.75 0.99
S4L 0.86 1.09 1.73 0.99 5.40 0.96 10.80 1.03
S4M 2.16 0.83 4.32 0.95 13.50 1.04 27.00 1.07
S4H 4.49 0.84 8.99 0.95 28.08 1.05 56.16 1.07
S5L 0.86 1.14 1.73 1.04 5.40 0.98 10.80 1.01
S5M 2.16 0.84 4.32 0.95 13.50 1.03 27.00 1.07
S5H 4.49 0.84 8.99 0.95 28.08 1.03 56.16 1.06
C1L 0.72 0.99 1.44 0.96 4.50 0.90 9.00 1.01
C1M 1.80 0.79 3.60 0.88 11.25 0.99 22.50 1.06
C1H 3.46 0.87 6.91 0.96 21.60 1.02 43.20 1.06
C2L 0.72 1.08 1.44 1.05 4.50 0.95 9.00 0.99
C2M 1.80 0.84 3.60 0.87 11.25 1.00 22.50 1.06
C2H 3.46 0.79 6.91 0.93 21.60 0.99 43.20 1.07
C3L 0.72 1.13 1.44 1.08 4.50 0.95 9.00 1.00
C3M 1.80 0.88 3.60 0.92 11.25 1.00 22.50 1.06
C3H 3.46 0.83 6.91 0.96 21.60 1.02 43.20 1.06
PC1 0.54 1.05 1.08 1.10 3.38 1.10 6.75 0.93
PC2L 0.72 1.12 1.44 1.04 4.50 0.93 9.00 1.02
PC2M 1.80 0.86 3.60 0.93 11.25 1.02 22.50 1.07
PC2H 3.46 0.83 6.91 0.94 21.60 1.04 43.20 1.07
RM1L 0.72 1.15 1.44 1.12 4.50 1.03 9.00 0.99
RM1M 1.80 0.89 3.60 0.89 11.25 1.00 22.50 1.05
RM2L 0.72 1.09 1.44 1.08 4.50 1.01 9.00 0.99
RM2M 1.80 0.85 3.60 0.86 11.25 0.99 22.50 1.06
1565
HazusMHTechnicalManual
RM2H 3.46 0.79 6.91 0.92 21.60 0.99 43.20 1.06
URML 0.54 1.07 1.08 1.13 3.38 1.16 6.75 1.01
URMM 1.26 0.97 2.52 0.91 7.88 0.98 15.75 1.04
MH 0.48 0.96 0.96 1.05 3.00 1.07 6.00 0.93
1566
Chapter5DirectPhysicalDamageGeneralBuildingStock
Table5.11dNonstructuralDriftSensitiveFragilityCurveParameters
PreCodeSeismicDesignLevel
Building MedianSpectralDisplacement(inches)andLogstandardDeviation(Beta)
Type Slight Moderate Extensive Complete
Median Beta Median Beta Median Beta Median Beta
W1 0.50 1.07 1.01 1.11 3.15 1.11 6.30 1.14
W2 0.86 1.06 1.73 1.00 5.40 0.93 10.80 1.01
S1L 0.86 0.90 1.73 0.87 5.40 0.91 10.80 1.02
S1M 2.16 0.80 4.32 0.92 13.50 0.99 27.00 1.06
S1H 4.49 0.79 8.99 0.89 28.08 1.00 56.16 1.07
S2L 0.86 1.06 1.73 0.97 5.40 0.96 10.80 1.04
S2M 2.16 0.80 4.32 0.90 13.50 1.02 27.00 1.06
S2H 4.49 0.79 8.99 0.89 28.08 0.99 56.16 1.06
S3 0.54 1.11 1.08 1.05 3.38 0.96 6.75 1.00
S4L 0.86 1.12 1.73 1.00 5.40 0.99 10.80 1.05
S4M 2.16 0.86 4.32 0.99 13.50 1.06 27.00 1.10
S4H 4.49 0.88 8.99 0.99 28.08 1.07 56.16 1.09
S5L 0.86 1.18 1.73 1.06 5.40 0.98 10.80 1.03
S5M 2.16 0.86 4.32 0.99 13.50 1.05 27.00 1.09
S5H 4.49 0.88 8.99 0.91 28.08 1.05 56.16 1.09
C1L 0.72 1.02 1.44 0.98 4.50 0.93 9.00 1.03
C1M 1.80 0.81 3.60 0.91 11.25 1.02 22.50 1.06
C1H 3.46 0.90 6.91 0.99 21.60 1.05 43.20 1.10
C2L 0.72 1.14 1.44 1.08 4.50 0.97 9.00 1.00
C2M 1.80 0.88 3.60 0.90 11.25 1.03 22.50 1.07
C2H 3.46 0.83 6.91 0.97 21.60 1.05 43.20 1.07
C3L 0.72 1.19 1.44 1.11 4.50 0.99 9.00 1.02
C3M 1.80 0.92 3.60 0.95 11.25 1.03 22.50 1.09
C3H 3.46 0.86 6.91 0.90 21.60 1.04 43.20 1.09
PC1 0.54 1.18 1.08 1.16 3.38 1.12 6.75 0.95
PC2L 0.72 1.16 1.44 1.06 4.50 0.96 9.00 1.02
PC2M 1.80 0.87 3.60 0.95 11.25 1.04 22.50 1.07
PC2H 3.46 0.87 6.91 0.99 21.60 1.06 43.20 1.08
RM1L 0.72 1.22 1.44 1.14 4.50 1.03 9.00 0.99
RM1M 1.80 0.93 3.60 0.92 11.25 1.02 22.50 1.07
RM2L 0.72 1.17 1.44 1.12 4.50 1.01 9.00 0.99
RM2M 1.80 0.89 3.60 0.90 11.25 1.01 22.50 1.07
1567
HazusMHTechnicalManual
RM2H 3.46 0.82 6.91 0.96 21.60 1.04 43.20 1.07
URML 0.54 1.21 1.08 1.23 3.38 1.23 6.75 1.03
URMM 1.26 0.99 2.52 0.95 7.88 0.99 15.75 1.06
MH 0.48 1.15 0.96 1.09 3.00 0.93 6.00 0.99

1568
Chapter5DirectPhysicalDamageGeneralBuildingStock
5.4.3.6 Nonstructural Damage - Acceleration-Sensitive Components

Table 5.12 summarizes the peak floor acceleration values used by the Methodology to define the
median values of fragility curves for acceleration-sensitive nonstructural components of
buildings. Nonstructural damage acceleration values are assumed to be the same for each model
building type, but to vary by seismic design level.

Table5.12PeakFloorAccelerationsUsedtoDefineMedianValuesofDamageto
NonstructuralAccelerationSensitiveComponents
SeismicDesign FloorAccelerationattheThresholdofNonstructuralDamage(g)
Level Slight Moderate Extensive Complete
HighCode 0.30 0.60 1.20 2.40
ModerateCode 0.25 0.50 1.00 2.00
LowCode 0.20 0.40 0.80 1.60
PreCode 0.20 0.40 0.80 1.60

The floor acceleration values are used directly as median values, assuming average upper-floor
demand is represented by response at the point of the push-over mode displacement.

The total variability of each damage state, |
NSAds
, is modeled by the combination of following
three contributors to nonstructural acceleration-sensitive damage variability:

- uncertainty in the damagestate threshold of nonstructural components
(|
M(NSAds)
=0.6,foralldamagestatesandbuildingtypes),
- variability in capacity (response) properties of the model building type that
contains the nonstructural components of interest (|
C(Au)
= 0.25 for Code
buildings,|
C(Au)
=0.30forPreCodebuildings),and
- variability in response of the model building type due to the spatial
variabilityofgroundmotiondemand(|
D(A)
=0.45and|
C(V)
=0.50).
Each of these three contributors to damage state variability is assumed to be lognormally
distributed random variables. Capacity and demand are dependent parameters and a convolution
process is used to derive combined capacity/demand variability of each nonstructural damage
state. Capacity/demand variability is then combined with damage state uncertainty, as described
in Section 5.4.3.3.
1569
HazusMHTechnicalManual

Tables 5.13a, 5.13b, 5.13c and 5.13d summarize median and lognormal standard deviation
(|
NSAds
) values for Slight, Moderate, Extensive and Complete nonstructural acceleration-sensitive
damage states for High-Code. Moderate-Code, Low-Code and Pre-Code buildings, respectively.
Median values are the same for all building types. Lognormal standard deviation values are
slightly different for each building type. Note that for the following tables, shaded boxes indicate
types that are not permitted by current seismic codes.

1570
Chapter5DirectPhysicalDamageGeneralBuildingStock
Table5.13aNonstructuralAccelerationSensitiveFragilityCurveParameters
HighCodeSeismicDesignLevel
Building MedianSpectralAcceleration(g)andLogstandardDeviation(Beta)
Type Slight Moderate Extensive Complete
Median Beta Median Beta Median Beta Median Beta
W1 0.30 0.73 0.60 0.68 1.20 0.68 2.40 0.68
W2 0.30 0.70 0.60 0.67 1.20 0.67 2.40 0.68
S1L 0.30 0.67 0.60 0.67 1.20 0.68 2.40 0.67
S1M 0.30 0.67 0.60 0.68 1.20 0.67 2.40 0.67
S1H 0.30 0.68 0.60 0.67 1.20 0.67 2.40 0.67
S2L 0.30 0.67 0.60 0.67 1.20 0.67 2.40 0.67
S2M 0.30 0.69 0.60 0.66 1.20 0.66 2.40 0.66
S2H 0.30 0.68 0.60 0.66 1.20 0.65 2.40 0.65
S3 0.30 0.68 0.60 0.67 1.20 0.67 2.40 0.67
S4L 0.30 0.68 0.60 0.68 1.20 0.67 2.40 0.67
S4M 0.30 0.67 0.60 0.65 1.20 0.66 2.40 0.66
S4H 0.30 0.67 0.60 0.66 1.20 0.65 2.40 0.65
S5L
S5M
S5H
C1L 0.30 0.68 0.60 0.68 1.20 0.67 2.40 0.67
C1M 0.30 0.68 0.60 0.68 1.20 0.66 2.40 0.66
C1H 0.30 0.66 0.60 0.66 1.20 0.66 2.40 0.66
C2L 0.30 0.69 0.60 0.67 1.20 0.66 2.40 0.64
C2M 0.30 0.70 0.60 0.65 1.20 0.65 2.40 0.65
C2H 0.30 0.68 0.60 0.66 1.20 0.65 2.40 0.65
C3L
C3M
C3H
PC1 0.30 0.74 0.60 0.67 1.20 0.67 2.40 0.64
PC2L 0.30 0.68 0.60 0.67 1.20 0.67 2.40 0.67
PC2M 0.30 0.68 0.60 0.65 1.20 0.66 2.40 0.66
PC2H 0.30 0.67 0.60 0.65 1.20 0.65 2.40 0.65
RM1L 0.30 0.70 0.60 0.67 1.20 0.67 2.40 0.63
RM1M 0.30 0.72 0.60 0.66 1.20 0.65 2.40 0.65
RM2L 0.30 0.70 0.60 0.66 1.20 0.67 2.40 0.64
RM2M 0.30 0.72 0.60 0.65 1.20 0.65 2.40 0.65
1571
HazusMHTechnicalManual
RM2H 0.30 0.70 0.60 0.65 1.20 0.65 2.40 0.65
URML
URMM
MH 0.30 0.65 0.60 0.67 1.20 0.67 2.40 0.67

1572
Chapter5DirectPhysicalDamageGeneralBuildingStock
Table5.13bNonstructuralAccelerationSensitiveFragilityCurveParameters
ModerateCodeSeismicDesignLevel
Building MedianSpectralAcceleration(g)andLogstandardDeviation(Beta)
Type Slight Moderate Extensive Complete
Median Beta Median Beta Median Beta Median Beta
W1 0.25 0.73 0.50 0.68 1.00 0.67 2.00 0.64
W2 0.25 0.68 0.50 0.67 1.00 0.68 2.00 0.68
S1L 0.25 0.67 0.50 0.66 1.00 0.67 2.00 0.67
S1M 0.25 0.66 0.50 0.67 1.00 0.67 2.00 0.67
S1H 0.25 0.66 0.50 0.68 1.00 0.68 2.00 0.68
S2L 0.25 0.66 0.50 0.66 1.00 0.68 2.00 0.68
S2M 0.25 0.66 0.50 0.65 1.00 0.65 2.00 0.65
S2H 0.25 0.65 0.50 0.65 1.00 0.65 2.00 0.65
S3 0.25 0.67 0.50 0.66 1.00 0.65 2.00 0.65
S4L 0.25 0.66 0.50 0.66 1.00 0.66 2.00 0.66
S4M 0.25 0.65 0.50 0.65 1.00 0.65 2.00 0.65
S4H 0.25 0.65 0.50 0.66 1.00 0.66 2.00 0.66
S5L
S5M
S5H
C1L 0.25 0.67 0.50 0.66 1.00 0.66 2.00 0.66
C1M 0.25 0.66 0.50 0.65 1.00 0.63 2.00 0.63
C1H 0.25 0.65 0.50 0.67 1.00 0.67 2.00 0.67
C2L 0.25 0.68 0.50 0.66 1.00 0.68 2.00 0.68
C2M 0.25 0.67 0.50 0.64 1.00 0.67 2.00 0.67
C2H 0.25 0.66 0.50 0.65 1.00 0.65 2.00 0.65
C3L
C3M
C3H
PC1 0.25 0.68 0.50 0.67 1.00 0.66 2.00 0.66
PC2L 0.25 0.66 0.50 0.66 1.00 0.65 2.00 0.65
PC2M 0.25 0.65 0.50 0.65 1.00 0.65 2.00 0.65
PC2H 0.25 0.64 0.50 0.65 1.00 0.65 2.00 0.65
RM1L 0.25 0.68 0.50 0.67 1.00 0.67 2.00 0.67
RM1M 0.25 0.67 0.50 0.64 1.00 0.67 2.00 0.67
RM2L 0.25 0.68 0.50 0.66 1.00 0.67 2.00 0.67
RM2M 0.25 0.67 0.50 0.64 1.00 0.67 2.00 0.67
1573
HazusMHTechnicalManual
RM2H 0.25 0.66 0.50 0.64 1.00 0.64 2.00 0.64
URML
URMM
MH 0.25 0.65 0.50 0.67 1.00 0.67 2.00 0.67

1574
Chapter5DirectPhysicalDamageGeneralBuildingStock
Table5.13cNonstructuralAccelerationSensitiveFragilityCurveParameters
LowCodeSeismicDesignLevel
Building MedianSpectralAcceleration(g)andLogstandardDeviation(Beta)
Type Slight Moderate Extensive Complete
Median Beta Median Beta Median Beta Median Beta
W1 0.20 0.71 0.40 0.68 0.80 0.66 1.60 0.66
W2 0.20 0.67 0.40 0.67 0.80 0.70 1.60 0.70
S1L 0.20 0.65 0.40 0.68 0.80 0.68 1.60 0.68
S1M 0.20 0.66 0.40 0.68 0.80 0.68 1.60 0.68
S1H 0.20 0.67 0.40 0.65 0.80 0.65 1.60 0.65
S2L 0.20 0.65 0.40 0.68 0.80 0.68 1.60 0.68
S2M 0.20 0.65 0.40 0.68 0.80 0.68 1.60 0.68
S2H 0.20 0.65 0.40 0.68 0.80 0.68 1.60 0.68
S3 0.20 0.65 0.40 0.68 0.80 0.68 1.60 0.68
S4L 0.20 0.65 0.40 0.68 0.80 0.68 1.60 0.68
S4M 0.20 0.65 0.40 0.68 0.80 0.68 1.60 0.68
S4H 0.20 0.65 0.40 0.68 0.80 0.68 1.60 0.68
S5L 0.20 0.65 0.40 0.68 0.80 0.67 1.60 0.67
S5M 0.20 0.64 0.40 0.68 0.80 0.67 1.60 0.67
S5H 0.20 0.65 0.40 0.68 0.80 0.68 1.60 0.68
C1L 0.20 0.65 0.40 0.68 0.80 0.68 1.60 0.68
C1M 0.20 0.65 0.40 0.68 0.80 0.68 1.60 0.68
C1H 0.20 0.67 0.40 0.67 0.80 0.67 1.60 0.67
C2L 0.20 0.66 0.40 0.67 0.80 0.66 1.60 0.66
C2M 0.20 0.64 0.40 0.66 0.80 0.65 1.60 0.65
C2H 0.20 0.64 0.40 0.66 0.80 0.66 1.60 0.66
C3L 0.20 0.65 0.40 0.67 0.80 0.66 1.60 0.66
C3M 0.20 0.64 0.40 0.67 0.80 0.66 1.60 0.66
C3H 0.20 0.64 0.40 0.67 0.80 0.67 1.60 0.67
PC1 0.20 0.66 0.40 0.66 0.80 0.66 1.60 0.66
PC2L 0.20 0.65 0.40 0.68 0.80 0.68 1.60 0.68
PC2M 0.20 0.64 0.40 0.68 0.80 0.68 1.60 0.68
PC2H 0.20 0.64 0.40 0.67 0.80 0.67 1.60 0.67
RM1L 0.20 0.66 0.40 0.67 0.80 0.64 1.60 0.64
RM1M 0.20 0.64 0.40 0.66 0.80 0.64 1.60 0.64
RM2L 0.20 0.66 0.40 0.67 0.80 0.64 1.60 0.64
RM2M 0.20 0.64 0.40 0.66 0.80 0.65 1.60 0.65
1575
HazusMHTechnicalManual
RM2H 0.20 0.64 0.40 0.66 0.80 0.66 1.60 0.66
URML 0.20 0.68 0.40 0.65 0.80 0.65 1.60 0.65
URMM 0.20 0.64 0.40 0.66 0.80 0.66 1.60 0.66
MH 0.20 0.65 0.40 0.67 0.80 0.67 1.60 0.67

1576
Chapter5DirectPhysicalDamageGeneralBuildingStock
Table5.13dNonstructuralAccelerationSensitiveFragilityCurveParameters
PreCodeSeismicDesignLevel
Building MedianSpectralAcceleration(g)andLogstandardDeviation(Beta)
Type Slight Moderate Extensive Complete
Median Beta Median Beta Median Beta Median Beta
W1 0.20 0.72 0.40 0.70 0.80 0.67 1.60 0.67
W2 0.20 0.66 0.40 0.67 0.80 0.65 1.60 0.65
S1L 0.20 0.66 0.40 0.68 0.80 0.68 1.60 0.68
S1M 0.20 0.66 0.40 0.68 0.80 0.68 1.60 0.68
S1H 0.20 0.68 0.40 0.68 0.80 0.68 1.60 0.68
S2L 0.20 0.65 0.40 0.68 0.80 0.68 1.60 0.68
S2M 0.20 0.65 0.40 0.68 0.80 0.68 1.60 0.68
S2H 0.20 0.65 0.40 0.67 0.80 0.67 1.60 0.67
S3 0.20 0.65 0.40 0.68 0.80 0.68 1.60 0.68
S4L 0.20 0.66 0.40 0.68 0.80 0.68 1.60 0.68
S4M 0.20 0.65 0.40 0.68 0.80 0.68 1.60 0.68
S4H 0.20 0.66 0.40 0.68 0.80 0.68 1.60 0.68
S5L 0.20 0.65 0.40 0.68 0.80 0.68 1.60 0.68
S5M 0.20 0.65 0.40 0.68 0.80 0.68 1.60 0.68
S5H 0.20 0.65 0.40 0.68 0.80 0.68 1.60 0.68
C1L 0.20 0.66 0.40 0.68 0.80 0.68 1.60 0.68
C1M 0.20 0.66 0.40 0.68 0.80 0.68 1.60 0.68
C1H 0.20 0.68 0.40 0.68 0.80 0.68 1.60 0.68
C2L 0.20 0.65 0.40 0.67 0.80 0.67 1.60 0.67
C2M 0.20 0.64 0.40 0.67 0.80 0.67 1.60 0.67
C2H 0.20 0.65 0.40 0.67 0.80 0.67 1.60 0.67
C3L 0.20 0.65 0.40 0.68 0.80 0.68 1.60 0.68
C3M 0.20 0.64 0.40 0.67 0.80 0.67 1.60 0.67
C3H 0.20 0.65 0.40 0.67 0.80 0.67 1.60 0.67
PC1 0.20 0.67 0.40 0.66 0.80 0.66 1.60 0.66
PC2L 0.20 0.65 0.40 0.68 0.80 0.68 1.60 0.68
PC2M 0.20 0.65 0.40 0.68 0.80 0.68 1.60 0.68
PC2H 0.20 0.66 0.40 0.67 0.80 0.67 1.60 0.67
RM1L 0.20 0.66 0.40 0.67 0.80 0.66 1.60 0.66
RM1M 0.20 0.64 0.40 0.66 0.80 0.65 1.60 0.65
RM2L 0.20 0.66 0.40 0.67 0.80 0.67 1.60 0.67
RM2M 0.20 0.64 0.40 0.66 0.80 0.66 1.60 0.66
1577
HazusMHTechnicalManual
RM2H 0.20 0.65 0.40 0.67 0.80 0.67 1.60 0.67
URML 0.20 0.69 0.40 0.65 0.80 0.65 1.60 0.65
URMM 0.20 0.64 0.40 0.66 0.80 0.66 1.60 0.66
MH 0.20 0.67 0.40 0.65 0.80 0.65 1.60 0.65

5.4.4 Structural Fragility Curves - Equivalent Peak Ground Acceleration

Structural damage functions are expressed in terms of an equivalent value of PGA (rather than
spectral displacement) for evaluation of buildings that are components of lifelines. Only
structural damage functions are developed based on PGA, since structural damage is considered
the most appropriate measure of damage for lifeline facilities. Similar methods could be used to
develop nonstructural damage functions based on PGA. In this case, capacity curves are not
necessary to estimate building response and PGA is used directly as the PESH input to building
fragility curves. This section develops equivalent-PGA fragility curves based on the structural
damage functions of Tables 5.9a - 5.9d and standard spectrum shape properties of Chapter 4.

Median values of equivalent-PGA fragility curves are based on median values of spectral
displacement of the damage state of interest and an assumed demand spectrum shape that relates
spectral response to PGA. As such, median values of equivalent PGA are very sensitive to the
shape assumed for the demand spectrum (i.e., PESH-input spectrum reduced for damping greater
than 5% of critical as described in Section 5.6.2.1). Spectrum shape is influenced by earthquake
source (i.e., WUS vs. CEUS attenuation functions), earthquake magnitude (e.g., large vs. small
magnitude events), distance from source to site, site conditions (e.g., soil vs. rock) and effective
damping which varies based on building properties and earthquake duration (e.g., Short,
Moderate or Long duration).

It is not practical to create equivalent-PGA fragility curves for all possible factors that influence
demand spectrum shape. Rather, equivalent-PGA fragility curves are developed for a single set
of spectrum shape factors (reference spectrum), and a formula is provided for modifying damage
state medians to approximate other spectrum shapes. The reference spectrum represents ground
shaking of a large-magnitude (i.e., M ~ 7.0) western United States (WUS) earthquake for soil
sites (e.g., Site Class D) at site-to-source distances of 15 km, or greater. The demand spectrum
based on these assumptions is scaled uniformly at each period such that the spectrum intersects
the building capacity curve at the spectral displacement of the median value of the damage state
of interest. The PGA of the scaled demand spectrum defines the median value of equivalent-PGA
fragility. Figure 5.6 illustrates this scaling and intersection process for a typical building capacity
curve and Slight, Moderate, Extensive and Complete structural damage states.

1578
Chapter5DirectPhysicalDamageGeneralBuildingStock
The total variability of each equivalent-PGA structural damage state, |
SPGA
, is modeled by the
combination of following two contributors to damage variability:

- uncertainty in the damagestate threshold of the structural system (|
M(SPGA)

=0.4forallbuildingtypesanddamagestates),
- variability in response due to the spatial variability of ground motion
demand(|
D(V)
=0.5forlongperiodspectralresponse).

Figure5.6DevelopmentofEquivalentPGAMedianDamageValues.
The two contributors to damage state variability are assumed to be lognormally distributed,
independent random variables and the total variability is simply the square-root-sum-of-the-
squares combination of individual variability terms (i.e., |
SPGA
=0.64). Tables 5.16a, 5.16b,
5.16c and 5.16d summarize median and lognormal standard deviation (|
SPGA
) values for Slight,
Moderate, Extensive and Complete PGA-based structural damage states for High-Code,
Moderate-Code, Low-Code and Pre-Code buildings, respectively.

The values given in Tables 5.16a through 5.16d are appropriate for use in the evaluation of
scenario earthquakes whose demand spectrum shape is based on, or similar to, large-magnitude,
WUS ground shaking at soil sites (reference spectrum shape). For evaluation of building damage
due to scenario earthquakes whose spectra are not similar to the reference spectrum shape,
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 1 2 3 4 5 6 7 8 9 10
Spectral Displacement (inches)
S
p
e
c
t
r
a
l

A
c
c
e
l
e
r
a
t
i
o
n

(
g
'
s
)
Capacity
Curve
Reference Demand Spectrum Shape
Large-Magnitude (i.e., M
W
= 7.0)
Western United States Earthquake
Soil-Site Conditions (i.e., Site Class D)
Site/Source Distance > 15 km
C
SD[M]
PGA[S]
S
M
E
SD[C] SD[E] SD[S]
PGA[M]
PGA[E]
PGA[C]
1579
HazusMHTechnicalManual
damage-state median parameters may be adjusted to better represent equivalent-PGA structural
fragility for the spectrum shape of interest. This adjustment is based on: (1) site condition (if
different from Site Class D) and (2) the ratio of long-period spectral response (i.e., S
A1
) to PGA
(if different from a value of 1.5, the ratio of S
A1
to PGA of the reference spectrum shape).
Damage-state variability is not adjusted assuming that the variability associated with ground
shaking (although different for different source/site conditions) when combined with the
uncertainty in damage-state threshold, is approximately the same for all demand spectrum shapes.

Tables 4.2 and 4.3 provide spectral acceleration response factors for WUS rock (Site Class B) and
CEUS rock (Site Class B) locations, respectively. These tables are based on the default WUS and
CEUS attenuation functions and describe response ratios, S
AS
/PGA and S
AS
/S
A1
, as a function of
distance and earthquake magnitude. Although both short-period response (S
AS
) and long-period
response (S
A1
) can influence building fragility, long-period response typically dominates building
fragility and is the parameter used to relate spectral demand to PGA. Spectral response factors
given in Tables 4.2 and 4.3 are combined to form ratios of PGA/S
A1
as given in Table 5.14 and
Table 5.15, respectively, for different earthquake magnitudes and source/site distances.

Table5.14SpectrumShapeRatio,R
PGA/SA1
WUSRock(SiteClassB)
ClosestDistanceto PGA/S
A1
givenMagnitude,M:
FaultRupture
s5 6 7 >8
s10km 3.8 2.1 1.5 0.85
20km 3.3 1.8 1.2 0.85
40km 2.9 1.6 1.05 0.80
>80km 3.2 1.7 1.0 0.75

Table5.15SpectrumShapeRatio,R
PGA/SA1
CEUSRock(SiteClassB)
Hypocentral PGA/S
A1
givenMagnitude,M:
Distance s5 6 7 >8
s10km 7.8 3.5 2.1 1.1
20km 8.1 3.1 2.1 1.7
40km 6.1 2.6 1.8 1.6
>80km 4.3 1.9 1.4 1.3
1580
Chapter5DirectPhysicalDamageGeneralBuildingStock

Equivalent-PGA medians specified in Tables 5.16a through 5.16d for the reference spectrum
shape are converted to medians representing other spectrum shapes using the ratios of Tables 5.14
and 5.15, the soil amplification factor, F
V
, and Equation (5-6):

PGA PGA R
F
ds R ds
PGA SA
V
=
|
\

|
.
| ,
/
.
1
15
(56)

where: PGA
ds
isthemedianPGAofstructuraldamagestate,ds,
PGA
R ds ,
isthemedianPGAofstructuraldamagestate,ds,asgiveninTables
513athrough513dforthereferencespectrumshape
R
PGA/SA1
isthespectrumshaperatio,giveninTables5.145.15,and
F
V
isthesoilamplificationfactor,giveninTable4.10

In general, implementation of Equation (5-6) requires information on earthquake magnitude and
source-to-site distance to estimate the spectrum shape ratio for rock sites, and 1-second period
spectral acceleration at the site (to estimate the soil amplification factor). Note that for Tables
5.16a through 5.16d, shaded boxes indicate types that are not permitted by current seismic codes.

1581
HazusMHTechnicalManual
Table5.16aEquivalentPGAStructuralFragility
HighCodeSeismicDesignLevel
Building MedianEquivalentPGA(g)andLogstandardDeviation(Beta)
Type Slight Moderate Extensive Complete
Median Beta Median Beta Median Beta Median Beta
W1 0.26 0.64 0.55 0.64 1.28 0.64 2.01 0.64
W2 0.26 0.64 0.56 0.64 1.15 0.64 2.08 0.64
S1L 0.19 0.64 0.31 0.64 0.64 0.64 1.49 0.64
S1M 0.14 0.64 0.26 0.64 0.62 0.64 1.43 0.64
S1H 0.10 0.64 0.21 0.64 0.52 0.64 1.31 0.64
S2L 0.24 0.64 0.41 0.64 0.76 0.64 1.46 0.64
S2M 0.14 0.64 0.27 0.64 0.73 0.64 1.62 0.64
S2H 0.11 0.64 0.22 0.64 0.65 0.64 1.60 0.64
S3 0.15 0.64 0.26 0.64 0.54 0.64 1.00 0.64
S4L 0.24 0.64 0.39 0.64 0.71 0.64 1.33 0.64
S4M 0.16 0.64 0.28 0.64 0.73 0.64 1.56 0.64
S4H 0.13 0.64 0.25 0.64 0.69 0.64 1.63 0.64
S5L
S5M
S5H
C1L 0.21 0.64 0.35 0.64 0.70 0.64 1.37 0.64
C1M 0.15 0.64 0.27 0.64 0.73 0.64 1.61 0.64
C1H 0.11 0.64 0.22 0.64 0.62 0.64 1.35 0.64
C2L 0.24 0.64 0.45 0.64 0.90 0.64 1.55 0.64
C2M 0.17 0.64 0.36 0.64 0.87 0.64 1.95 0.64
C2H 0.12 0.64 0.29 0.64 0.82 0.64 1.87 0.64
C3L
C3M
C3H
PC1 0.20 0.64 0.35 0.64 0.72 0.64 1.25 0.64
PC2L 0.24 0.64 0.36 0.64 0.69 0.64 1.23 0.64
PC2M 0.17 0.64 0.29 0.64 0.67 0.64 1.51 0.64
PC2H 0.12 0.64 0.23 0.64 0.63 0.64 1.49 0.64
RM1L 0.30 0.64 0.46 0.64 0.93 0.64 1.57 0.64
RM1M 0.20 0.64 0.37 0.64 0.81 0.64 1.90 0.64
RM2L 0.26 0.64 0.42 0.64 0.87 0.64 1.49 0.64
RM2M 0.17 0.64 0.33 0.64 0.75 0.64 1.83 0.64
1582
Chapter5DirectPhysicalDamageGeneralBuildingStock
RM2H 0.12 0.64 0.24 0.64 0.67 0.64 1.78 0.64
URML
URMM
MH 0.11 0.64 0.18 0.64 0.31 0.64 0.60 0.64

1583
HazusMHTechnicalManual
Table5.16bEquivalentPGAStructuralFragility
ModerateCodeSeismicDesignLevel
Building MedianEquivalentPGA(g)andLogstandardDeviation(Beta)
Type Slight Moderate Extensive Complete
Median Beta Median Beta Median Beta Median Beta
W1 0.24 0.64 0.43 0.64 0.91 0.64 1.34 0.64
W2 0.20 0.64 0.35 0.64 0.64 0.64 1.13 0.64
S1L 0.15 0.64 0.22 0.64 0.42 0.64 0.80 0.64
S1M 0.13 0.64 0.21 0.64 0.44 0.64 0.82 0.64
S1H 0.10 0.64 0.18 0.64 0.39 0.64 0.78 0.64
S2L 0.20 0.64 0.26 0.64 0.46 0.64 0.84 0.64
S2M 0.14 0.64 0.22 0.64 0.53 0.64 0.97 0.64
S2H 0.11 0.64 0.19 0.64 0.49 0.64 1.02 0.64
S3 0.13 0.64 0.19 0.64 0.33 0.64 0.60 0.64
S4L 0.19 0.64 0.26 0.64 0.41 0.64 0.78 0.64
S4M 0.14 0.64 0.22 0.64 0.51 0.64 0.92 0.64
S4H 0.12 0.64 0.21 0.64 0.51 0.64 0.97 0.64
S5L
S5M
S5H
C1L 0.16 0.64 0.23 0.64 0.41 0.64 0.77 0.64
C1M 0.13 0.64 0.21 0.64 0.49 0.64 0.89 0.64
C1H 0.11 0.64 0.18 0.64 0.41 0.64 0.74 0.64
C2L 0.18 0.64 0.30 0.64 0.49 0.64 0.87 0.64
C2M 0.15 0.64 0.26 0.64 0.55 0.64 1.02 0.64
C2H 0.12 0.64 0.23 0.64 0.57 0.64 1.07 0.64
C3L
C3M
C3H
PC1 0.18 0.64 0.24 0.64 0.44 0.64 0.71 0.64
PC2L 0.18 0.64 0.25 0.64 0.40 0.64 0.74 0.64
PC2M 0.15 0.64 0.21 0.64 0.45 0.64 0.86 0.64
PC2H 0.12 0.64 0.19 0.64 0.46 0.64 0.90 0.64
RM1L 0.22 0.64 0.30 0.64 0.50 0.64 0.85 0.64
RM1M 0.18 0.64 0.26 0.64 0.51 0.64 1.03 0.64
RM2L 0.20 0.64 0.28 0.64 0.47 0.64 0.81 0.64
RM2M 0.16 0.64 0.23 0.64 0.48 0.64 0.99 0.64
1584
Chapter5DirectPhysicalDamageGeneralBuildingStock
RM2H 0.12 0.64 0.20 0.64 0.48 0.64 1.01 0.64
URML
URMM
MH 0.11 0.64 0.18 0.64 0.31 0.64 0.60 0.64

1585
HazusMHTechnicalManual
Table5.16cEquivalentPGAStructuralFragility
LowCodeSeismicDesignLevel
Building MedianEquivalentPGA(g)andLogstandardDeviation(Beta)
Type Slight Moderate Extensive Complete
Median Beta Median Beta Median Beta Median Beta
W1 0.20 0.64 0.34 0.64 0.61 0.64 0.95 0.64
W2 0.14 0.64 0.23 0.64 0.48 0.64 0.75 0.64
S1L 0.12 0.64 0.17 0.64 0.30 0.64 0.48 0.64
S1M 0.12 0.64 0.18 0.64 0.29 0.64 0.49 0.64
S1H 0.10 0.64 0.15 0.64 0.28 0.64 0.48 0.64
S2L 0.13 0.64 0.17 0.64 0.30 0.64 0.50 0.64
S2M 0.12 0.64 0.18 0.64 0.35 0.64 0.58 0.64
S2H 0.11 0.64 0.17 0.64 0.36 0.64 0.63 0.64
S3 0.10 0.64 0.13 0.64 0.20 0.64 0.38 0.64
S4L 0.13 0.64 0.16 0.64 0.26 0.64 0.46 0.64
S4M 0.12 0.64 0.17 0.64 0.31 0.64 0.54 0.64
S4H 0.12 0.64 0.17 0.64 0.33 0.64 0.59 0.64
S5L 0.13 0.64 0.17 0.64 0.28 0.64 0.45 0.64
S5M 0.11 0.64 0.18 0.64 0.34 0.64 0.53 0.64
S5H 0.10 0.64 0.18 0.64 0.35 0.64 0.58 0.64
C1L 0.12 0.64 0.15 0.64 0.27 0.64 0.45 0.64
C1M 0.12 0.64 0.17 0.64 0.32 0.64 0.54 0.64
C1H 0.10 0.64 0.15 0.64 0.27 0.64 0.44 0.64
C2L 0.14 0.64 0.19 0.64 0.30 0.64 0.52 0.64
C2M 0.12 0.64 0.19 0.64 0.38 0.64 0.63 0.64
C2H 0.11 0.64 0.19 0.64 0.38 0.64 0.65 0.64
C3L 0.12 0.64 0.17 0.64 0.26 0.64 0.44 0.64
C3M 0.11 0.64 0.17 0.64 0.32 0.64 0.51 0.64
C3H 0.09 0.64 0.16 0.64 0.33 0.64 0.53 0.64
PC1 0.13 0.64 0.17 0.64 0.25 0.64 0.45 0.64
PC2L 0.13 0.64 0.15 0.64 0.24 0.64 0.44 0.64
PC2M 0.11 0.64 0.16 0.64 0.31 0.64 0.52 0.64
PC2H 0.11 0.64 0.16 0.64 0.31 0.64 0.55 0.64
RM1L 0.16 0.64 0.20 0.64 0.29 0.64 0.54 0.64
RM1M 0.14 0.64 0.19 0.64 0.35 0.64 0.63 0.64
RM2L 0.14 0.64 0.18 0.64 0.28 0.64 0.51 0.64
RM2M 0.12 0.64 0.17 0.64 0.34 0.64 0.60 0.64
1586
Chapter5DirectPhysicalDamageGeneralBuildingStock
RM2H 0.11 0.64 0.17 0.64 0.35 0.64 0.62 0.64
URML 0.14 0.64 0.20 0.64 0.32 0.64 0.46 0.64
URMM 0.10 0.64 0.16 0.64 0.27 0.64 0.46 0.64
MH 0.11 0.64 0.18 0.64 0.31 0.64 0.60 0.64

Table5.16dEquivalentPGAStructuralFragility
PreCodeSeismicDesignLevel
Building MedianEquivalentPGA(g)andLogstandardDeviation(Beta)
Type Slight Moderate Extensive Complete
Median Beta Median Beta Median Beta Median Beta
W1 0.18 0.64 0.29 0.64 0.51 0.64 0.77 0.64
W2 0.12 0.64 0.19 0.64 0.37 0.64 0.60 0.64
S1L 0.09 0.64 0.13 0.64 0.22 0.64 0.38 0.64
S1M 0.09 0.64 0.14 0.64 0.23 0.64 0.39 0.64
S1H 0.08 0.64 0.12 0.64 0.22 0.64 0.38 0.64
S2L 0.11 0.64 0.14 0.64 0.23 0.64 0.39 0.64
S2M 0.10 0.64 0.14 0.64 0.28 0.64 0.47 0.64
S2H 0.09 0.64 0.13 0.64 0.29 0.64 0.50 0.64
S3 0.08 0.64 0.10 0.64 0.16 0.64 0.30 0.64
S4L 0.10 0.64 0.13 0.64 0.20 0.64 0.36 0.64
S4M 0.09 0.64 0.13 0.64 0.25 0.64 0.43 0.64
S4H 0.09 0.64 0.14 0.64 0.27 0.64 0.47 0.64
S5L 0.11 0.64 0.14 0.64 0.22 0.64 0.37 0.64
S5M 0.09 0.64 0.14 0.64 0.28 0.64 0.43 0.64
S5H 0.08 0.64 0.14 0.64 0.29 0.64 0.46 0.64
C1L 0.10 0.64 0.12 0.64 0.21 0.64 0.36 0.64
C1M 0.09 0.64 0.13 0.64 0.26 0.64 0.43 0.64
C1H 0.08 0.64 0.12 0.64 0.21 0.64 0.35 0.64
C2L 0.11 0.64 0.15 0.64 0.24 0.64 0.42 0.64
C2M 0.10 0.64 0.15 0.64 0.30 0.64 0.50 0.64
C2H 0.09 0.64 0.15 0.64 0.31 0.64 0.52 0.64
C3L 0.10 0.64 0.14 0.64 0.21 0.64 0.35 0.64
C3M 0.09 0.64 0.14 0.64 0.25 0.64 0.41 0.64
C3H 0.08 0.64 0.13 0.64 0.27 0.64 0.43 0.64
PC1 0.11 0.64 0.14 0.64 0.21 0.64 0.35 0.64
PC2L 0.10 0.64 0.13 0.64 0.19 0.64 0.35 0.64
1587
HazusMHTechnicalManual
PC2M 0.09 0.64 0.13 0.64 0.24 0.64 0.42 0.64
PC2H 0.09 0.64 0.13 0.64 0.25 0.64 0.43 0.64
RM1L 0.13 0.64 0.16 0.64 0.24 0.64 0.43 0.64
RM1M 0.11 0.64 0.15 0.64 0.28 0.64 0.50 0.64
RM2L 0.12 0.64 0.15 0.64 0.22 0.64 0.41 0.64
RM2M 0.10 0.64 0.14 0.64 0.26 0.64 0.47 0.64
RM2H 0.09 0.64 0.13 0.64 0.27 0.64 0.50 0.64
URML 0.13 0.64 0.17 0.64 0.26 0.64 0.37 0.64
URMM 0.09 0.64 0.13 0.64 0.21 0.64 0.38 0.64
MH 0.08 0.64 0.11 0.64 0.18 0.64 0.34 0.64

1588
Chapter5DirectPhysicalDamageGeneralBuildingStock
5.5 Building Damage Due to Ground Failure

5.5.1 Overview

Building damage is characterized by four damage states (i.e., Slight, Moderate, Extensive and
Complete). These four states are simplified for ground failure to include only one combined
Extensive/Complete damage state. In essence, buildings are assumed to be either undamaged or
severely damaged due to ground failure. In fact, Slight or Moderate damage can occur due to
ground failure, but the likelihood of this damage is considered to be small (relative to ground
shaking damage) and tacitly included in predictions of Slight or Moderate damage due to ground
shaking.

Given the earthquake demand in terms of permanent ground deformation (PGD), the probability
of being in the Extensive/Complete damage state is estimated using fragility curves of a form
similar to those used to estimate shaking damage. Separate fragility curves distinguish between
ground failure due to lateral spreading and ground failure due to ground settlement, and between
shallow and deep foundations.

5.5.2 FragilityCurvesPeakGroundDisplacement

There is no available relationship between the likelihood of Extensive/Complete damage of
buildings and PGD. Engineering judgment is used to develop a set of assumptions, which define
building fragility. These assumptions are shown in Table 5.17 for buildings with shallow
foundations (e.g., spread footings).

Table5.17BuildingDamageRelationshiptoPGDShallowFoundations
P E or C|PGD SettlementPGD
(inches)
LateralSpreadPGD
(inches)
0.1 2 12
0.5(median) 10 60

The above assumptions are based on the expectation that about 10 (i.e., 8 Extensive damage, 2
Complete damage) out of 100 buildings on spread footings would be severely damaged for 2
1589
HazusMHTechnicalManual
inches of settlement PGD or 12 inches of lateral spread PGD, and that about 50 (i.e., 40 Extensive
damage, 10 Complete damage) out of 100 buildings on spread footings would be severely
damaged for 10 inches of settlement PGD or 60 inches of lateral spread PGD. Lateral spread is
judged to require significantly more PGD to effect severe damage than ground settlement. Many
buildings in lateral spread areas are expected to move with the spread, but not to be severely
damaged until the spread becomes quite significant.

Median PGD values given in the Table 5.17 are used with a lognormal standard deviation value
of |
PGD
= 1.2 to estimate P[E or C|PGD] for buildings on shallow foundations or buildings of
unknown foundation type. The value of |
PGD
= 1.2 is based on the factor of 5 between the PGD
values at the 10 and 50 percentile levels.

No attempt is made to distinguish damage based on building type, since model building
descriptions do not include foundation type. Foundation type is critical to PGD performance and
buildings on deep foundations (e.g., piles) perform much better than buildings on spread footings,
if the ground settles. When the building is known to be supported by a deep foundation, the
probability of Extensive or Complete damage is reduced by a factor of 10 from that predicted for
settlement-induced damage of the same building on a shallow foundation. Deep foundations will
improve building performance by only a limited amount, if ground spreads laterally. When the
building is known to be supported by a deep foundation, the probability of Extensive or Complete
damage is reduced by a factor of 2 from that predicted for spread-induced damage of the same
building on a shallow foundation.

5.6 Evaluation of Building Damage

5.6.1 Overview
During an earthquake, the building may be damaged either by ground shaking, ground failure, or
both. Buildings are evaluated separately for the two modes of failure the resulting damage-state
probabilities are combined for evaluation of loss.

5.6.2 Damage Due to Ground Shaking

This section describes the process of developing damage state probabilities based on structural
and nonstructural fragility curves, model building capacity curves and a demand spectrum.
Building response (e.g., peak displacement) is determined by the intersection of the demand
spectrum and the building capacity curve. The demand spectrum is based on the PESH input
1590
Chapter5DirectPhysicalDamageGeneralBuildingStock
spectrum reduced for effective damping (when effective damping exceeds the 5% damping level
of the PESH input spectrum).

5.6.2.1DemandSpectrumReductionforEffectiveDamping

The elastic response spectra provided as a PESH input apply only to buildings that remain elastic
during the entire ground shaking time history and have elastic damping values equal to 5% of
critical. This is generally not true on both accounts. Therefore, two modifications are made to
elastic response spectra: (a) demand spectra are modified for buildings with elastic damping not
equal to 5%, and (b) demand spectra are modified for the hysteretic energy dissipated by
buildings pushed beyond their elastic limits. Modifications are represented by reduction factors
by which the spectral ordinates are divided to obtain the damped demand spectra.

Extensive work has been published in the past two decades on the effect of damping and/or
energy dissipation on spectral demand. The Methodology reduces demand spectra for effective
damping greater than 5% based on statistically-based formulas of Newmark and Hall (1982).
Other methods are available for estimating spectral reduction factors based on statistics relating
reduction to ductility demand. It is believed that both methods yield the same results for most
practical purposes (FEMA 273). Newmark and Hall provide formulas for construction of elastic
response spectra at different damping ratios, B (expressed as a percentage). These formulas
represent all site classes (soil types) distinguishing between domains of constant acceleration and
constant velocity. Ratios of these formulas are used to develop an acceleration-domain (short-
period) reduction factor, R
A
, and a velocity-domain (1-second spectral acceleration) reduction
factor, R
V
, for modification of 5%-damped, elastic response spectra (PESH input). These
reduction factors are based on effective damping, B
eff
, as given in Equations (5-7) and (5-8)
below:

( ) R
A eff
= 212 321 068 . . . ln( ) B (57)

( ) R
V eff
= 165 231 041 . . . ln( ) B (58)

for which effective damping is defined as the sum of elastic damping, B


E
, and hysteretic damping,
B
H
:

B B B
eff E H
= + (59)

Elastic damping, B
E
, is dependent on structure type and is based on the recommendations of
Newmark & Hall for materials at or just below their yield point. Hysteretic damping, B
H
, is
1591
HazusMHTechnicalManual
dependent on the amplitude of response and is based on the area enclosed by the hysteresis loop,
considering potential degradation of energy-absorption capacity of the structure during cyclic
earthquake load. Effective damping, B
eff
, is also a function of the amplitude of response (e.g.,
peak displacement), as expressed in Equation (5-10):

B B
Area
D A
eff E
= +

|
\

|
.
|
k
t 2
(510)

where: B
E
istheelastic(preyield)dampingofthemodelbuildingtype
Area is the area enclosed by the hysteresis loop, as defined by a
symmetrical pushpull of the building capacity curve up to peak
positiveandnegativedisplacements,D
D isthepeakdisplacementresponseofthepushovercurve,
A isthepeakaccelerationresponseatpeakdisplacement,D
k is a degradation factor that defines the effective amount of hysteretic
damping as a function of earthquake duration, as specified in Table
5.18.

Table5.18DegradationFactor(k)asaFunctionofShort,ModerateandLongEarthquake
BuildingType High-CodeDesign Moderate-CodeDesign Low-CodeDesign Pre-CodeDesign
No. Label Short Moderate Long Short Moderate Long Short Moderate Long Short Moderate Long
1 W1 1.00 0.80 0.50 0.90 0.60 0.30 0.70 0.40 0.20 0.50 0.30 0.10
2 W2 0.90 0.60 0.40 0.80 0.40 0.20 0.60 0.30 0.10 0.40 0.20 0.00
3 S1L 0.90 0.60 0.40 0.80 0.40 0.20 0.60 0.30 0.10 0.40 0.20 0.00
4 S1M 0.90 0.60 0.40 0.80 0.40 0.20 0.60 0.30 0.10 0.40 0.20 0.00
5 S1H 0.90 0.60 0.40 0.80 0.40 0.20 0.60 0.30 0.10 0.40 0.20 0.00
6 S2L 0.70 0.50 0.30 0.60 0.40 0.20 0.50 0.30 0.10 0.40 0.20 0.00
7 S2M 0.70 0.50 0.30 0.60 0.40 0.20 0.50 0.30 0.10 0.40 0.20 0.00
8 S2H 0.70 0.50 0.30 0.60 0.40 0.20 0.50 0.30 0.10 0.40 0.20 0.00
9 S3 0.70 0.50 0.30 0.60 0.40 0.20 0.50 0.30 0.10 0.40 0.20 0.00
10 S4L 0.70 0.50 0.30 0.60 0.40 0.20 0.50 0.30 0.10 0.40 0.20 0.00
11 S4M 0.70 0.50 0.30 0.60 0.40 0.20 0.50 0.30 0.10 0.40 0.20 0.00
12 S4H 0.70 0.50 0.30 0.60 0.40 0.20 0.50 0.30 0.10 0.40 0.20 0.00
13 S5L 0.50 0.30 0.10 0.50 0.30 0.10 0.50 0.30 0.10 0.40 0.20 0.00
14 S5M 0.50 0.30 0.10 0.50 0.30 0.10 0.50 0.30 0.10 0.40 0.20 0.00
15 S5H 0.50 0.30 0.10 0.50 0.30 0.10 0.50 0.30 0.10 0.40 0.20 0.00
16 C1L 0.90 0.60 0.40 0.80 0.40 0.20 0.60 0.30 0.10 0.40 0.20 0.00
17 C1M 0.90 0.60 0.40 0.80 0.40 0.20 0.60 0.30 0.10 0.40 0.20 0.00
18 C1H 0.90 0.60 0.40 0.80 0.40 0.20 0.60 0.30 0.10 0.40 0.20 0.00
19 C2L 0.90 0.60 0.40 0.80 0.40 0.20 0.60 0.30 0.10 0.40 0.20 0.00
20 C2M 0.90 0.60 0.40 0.80 0.40 0.20 0.60 0.30 0.10 0.40 0.20 0.00
21 C2H 0.90 0.60 0.40 0.80 0.40 0.20 0.60 0.30 0.10 0.40 0.20 0.00
22 C3L 0.50 0.30 0.10 0.50 0.30 0.10 0.50 0.30 0.10 0.40 0.20 0.00
23 C3M 0.50 0.30 0.10 0.50 0.30 0.10 0.50 0.30 0.10 0.40 0.20 0.00
24 C3H 0.50 0.30 0.10 0.50 0.30 0.10 0.50 0.30 0.10 0.40 0.20 0.00
25 PC1 0.70 0.50 0.30 0.60 0.40 0.20 0.50 0.30 0.10 0.40 0.20 0.00
26 PC2L 0.70 0.50 0.30 0.60 0.40 0.20 0.50 0.30 0.10 0.40 0.20 0.00
27 PC2M 0.70 0.50 0.30 0.60 0.40 0.20 0.50 0.30 0.10 0.40 0.20 0.00
28 PC2H 0.70 0.50 0.30 0.60 0.40 0.20 0.50 0.30 0.10 0.40 0.20 0.00
29 RM1L 0.90 0.60 0.40 0.80 0.40 0.20 0.60 0.30 0.10 0.40 0.20 0.00
30 RM1M 0.90 0.60 0.40 0.80 0.40 0.20 0.60 0.30 0.10 0.40 0.20 0.00
31 RM2L 0.90 0.60 0.40 0.80 0.40 0.20 0.60 0.30 0.10 0.40 0.20 0.00
32 RM2M 0.90 0.60 0.40 0.80 0.40 0.20 0.60 0.30 0.10 0.40 0.20 0.00
33 RM2H 0.90 0.60 0.40 0.80 0.40 0.20 0.60 0.30 0.10 0.40 0.20 0.00
34 URML 0.50 0.30 0.10 0.50 0.30 0.10 0.50 0.30 0.10 0.40 0.20 0.00
35 URMM 0.50 0.30 0.10 0.50 0.30 0.10 0.50 0.30 0.10 0.40 0.20 0.00
36 MH 0.80 0.40 0.20 0.80 0.40 0.20 0.80 0.40 0.20 0.60 0.30 0.10

1592
Chapter5DirectPhysicalDamageGeneralBuildingStock
Duration


The Methodology recognizes the importance of the duration of ground shaking on building
response by reducing effective damping (i.e., k factors) as a function of shaking duration.
Shaking duration is described qualitatively as either Short, Moderate or Long, and is assumed to
be a function of earthquake magnitude (although proximity to fault rupture also influences the
duration of ground shaking). For scenario earthquakes of magnitude M s 5.5, effective damping
is based on the assumption of ground shaking of Short duration. For scenario earthquakes of
magnitude M > 7.5, effective damping is based on the assumption of ground shaking of Long
duration. Effective damping is based on the assumption of Moderate duration for all other
earthquake magnitudes (including probabilistic, or other, analyses of unknown magnitude).


Construction of Demand Spectra
Demand spectral acceleration, S
A
[T], in units of acceleration (g) is defined by Equation (5-11a) at
short periods (acceleration domain), Equation (5-11b) at long periods (velocity domain) and
Equation (5-11c) at very long periods (displacement domain).

Atshortperiods,0<TsT
AV|
:
( ) ( ) ( )
S T] S R B S B
A ASi A eff ASi eff
[ [ ] . . . ln = = 212 321 068 (511a)
Atlongperiods,T
AV|
<TsT
VD
:
( ) ( ) ( )
S T]
S
T
R B
S
T
B
A
A i
V eff
A i
eff
[ [ ] . . . ln =
|
\

|
.
|
=
|
\

|
.
|

1 1
165 231 041 (511b)
Atverylongperiods,T>T
VD
:
( ) ( ) ( )
S T]
S T
T
R B
S T
T
B
A
A i VD
V TVD
A i VD
TVD
[ [ ] . . . ln =
|
\

|
.
|
=
|
\

|
.
|

1
2
1
2
165 231 041 (511c)
where: S
ASi
isthe5%damped,shortperiodspectralaccelerationforSiteClassi (in
unitsofg),asdefinedbyEquation(45),
S
A1i
isthe5%damped,1secondperiodspectralaccelerationforSiteClassi
(unitsofg),asdefinedbyEquation(46),times1second,
1593
HazusMHTechnicalManual
T
AVi
is the transition period between 5%damped constant spectral
acceleration and 5%damped constant spectral velocity for Site Class i
(sec.),asdefinedbyEquation(47),
B
TVD
isthevalueofeffectivedampingatthetransitionperiod,T
VD
,and
B
TAVB
isthevalueofeffectivedampingatthetransitionperiod,T
AVB
.
The transition period, T
AVB
, between acceleration and velocity domains is a function of the
effective damping at this period, as defined by Equation (5-12). The transition period, T
VD
,
between velocity and displacement domains is independent of effective damping, as defined by
Equation (4-4).

| |
| |
( ) ( )
( ) ( )
T T
R B
R B
T
B
B
AVB AVi
A TAVB
V TAVB
AVi
TAVB
TAVB
=
|
\

|
.
| =

|
\

|
.
|
|
212 321 068
165 231 041
. . . ln
. . . ln
(512)

Demand spectral displacement, S
D
[T], in inches, is based on S
A
[T], in units of g, as given on
Equation (5-13):

S T] S T] T
D A
[ . [ = 98
2
(513)

Figure 5.7 shows typical demand spectra (spectral acceleration plotted as a function of spectral
displacement) for three demand levels. These three demand levels represent Short (k =0.80),
Moderate (k =0.40) and Long (k =0.20) duration ground shaking, respectively. Also shown in
the figure is the building capacity curve of a low-rise building of Moderate-Code seismic design
that was used to estimate effective damping. The intersection of the capacity curve with each of
the three demand spectra illustrates the significance of duration (damping) on building response.

1594
Chapter5DirectPhysicalDamageGeneralBuildingStock
Figure5.7ExampleDemandSpectraModerateCodeBuilding
(M=7.0at20km,WUS,SiteClassE).

5.6.2.2 Damage State Probability

Structural and nonstructural fragility curves are evaluated for spectral displacement and spectral
acceleration defined by the intersection of the capacity and demand curves. Each of these curves
describes the cumulative probability of being in, or exceeding, a particular damage state.
Nonstructural components (both drift- and acceleration-sensitive components) may, in some
cases, be dependent on the structural damage state (e.g., Complete structural damage may cause
Complete nonstructural damage). The Methodology assumes nonstructural damage states to be
independent of structural damage states. Cumulative probabilities are differenced to obtain
discrete probabilities of being in each of the five damage states. This process is shown
schematically in Figure 5.8.

0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 1 2 3 4 5 6 7 8 9 10
Spectral Displacement (inches)
S
p
e
c
t
r
a
l

A
c
c
e
l
e
r
a
t
i
o
n

(
g
'
s
)
PESH Spectrum - 5% Damping
Demand Spectrum - Long Duration
Demand Spectrum - Moderate Duration
Demand Spectrum - Short Duration
Building Capacity Curve
0.3 sec.
1.0 sec.
1595
HazusMHTechnicalManual
Figure5.8ExampleBuildingDamageEstimationProcess.
It is also meaningful to interpret damage probabilities as the fraction of all buildings (of the same
type) that would be in the particular damage state of interest. For example, a 30% probability of
Moderate damage may also be thought of as 30 out of 100 buildings (of the same type) being in
the Moderate damage state.

5.6.3 CombinedDamageDuetoGroundFailureandGroundShaking
This section describes the combination of damage state probabilities due to ground failure
(Section 5.5.2) and ground shaking (Section 5.6.2.2). It is assumed that damage due to ground
shaking (GS) is independent of damage due to ground failure (GF). Ground failure tends to cause
severe damage to buildings and is assumed to contribute only to Extensive and Complete damage
states (refer to Section 5.5.1). These assumptions are described by the following formulas:

S
C
C
S S
C
M
E
M
E
M
E
1596
Chapter5DirectPhysicalDamageGeneralBuildingStock

| | | |
P DS S P DS E
GF GF
> = > (514)


| | | |
P DS M P DS E
GF GF
> = > (515)


| | | |
P DS C 0.2 P DS E
GF GF
> = > (516)
The damage state probability (probability of being in or exceeding a given damage state) for GF
is assumed to be the maximum of the three types of ground failure (liquefaction, landsliding, and
spread). Thus, the combined (due to occurrence of GF or GS) probabilities of being in or
exceeding given damage states are:


| | | | | | | | | |
P DS S P DS S P DS S P DS S P DS S
COMB GF GS GF GS
> = > + > > > (5-17)

| | | | | | | | | |
P DS M P DS M P DS M P DS M P DS M
COMB GF GS GF GS
> = > + > > > (5-18)

| | | | | | | | | |
P DS E P DS E P DS E P DS E P DS E
COMB GF GS GF GS
> = > + > > > (5-19)

| | | | | | | | | |
P DS C P DS C P DS C P DS C P DS C
COMB GF GS GF GS
> = > + > > > (5-20)


where DS is damage state, and the symbols: S, M, E, and C stand for Slight, Moderate,
Extensive, and Complete damage, respectively. COMB indicates the combined probability for
the damage state due to occurrence of ground failure or ground shaking. Note that the following
condition must always be true:

| | | | | | | |
1 P DS S P DS M P DS E P DS C
COMB COMB COMB COMB
> > > > > > > > (5-21)
1597
HazusMHTechnicalManual


The discrete probabilities (probabilities of being in a given damage state) are given as:



| | | |
P DS C P DS C
COMB COMB
= = > (5-22)


| | | | | |
P DS E P DS E P DS C
COMB COMB COMB
= = > > (5-23)


| | | | | |
P DS M P DS M P DS E
COMB COMB COMB
= = > > (5-24)


| | | | | |
P DS S P DS S P DS M
COMB COMB COMB
= = > > (5-25)


| | | |
P DS None 1 P DS S
COMB COMB
= = > (5-26)

5.6.4 CombinedDamagetoOccupancyClasses

The damage state probabilities for model building types (as estimated from Section 5.6.3) are
combined to yield the damage state probabilities of the occupancy classes to which they belong.
For each damage state, the probability of damage to each model building type is weighted
according to the fraction of the total floor area of that model building type and summed over all
building types. This is expressed in equation form:

POSTR PMBTSTR
FA
FA
ds,i ds,j
i,j
i
j 1
36
=

=
(5-27)

1598
Chapter5DirectPhysicalDamageGeneralBuildingStock
where PMBTSTR
ds,j
is the probability of the model building type j being in damage state ds.
POSTR
ds,i
is the probability of occupancy class i being in damage state ds. FA
i,j
indicates the
floor area of model building type j in occupancy class i, and FA
i
denotes the total floor area of the
occupancy class i (refer to Chapter 3 for floor area distributions of model building types by
occupancy class). Similarly, the damage-state probabilities for nonstructural components can be
estimated.

PONSD PMBTNSD
FA
FA
ds,i ds,j
i,j
i
j 1
36
=

=
(5-28)


PONSA PMBTNSA
FA
FA
ds,i ds,j
i,j
i
j 1
36
=

=
(5-29)

where PMBTNSD
ds,j
and PMBTNSA
ds,j
refer to the probabilities of model building type j being
in nonstructural drift- and acceleration-sensitive damage state ds, respectively; and PONSD
ds,i

and PONSA
ds,i
refer to the probabilities of the occupancy class i being the nonstructural drift-
sensitive and acceleration-sensitive damage state, ds, respectively. These occupancy class
probabilities are used in Chapter 15 to estimate direct economic loss.

5.7 GuidanceforExpertUsers
This section provides guidance for users who are seismic/structural experts interested in
modifying the building damage functions supplied with the methodology. This section also
provides the expert user with guidance regarding the selection of the appropriate mix of design
levels for the region of interest.


5.7.1 SelectionofRepresentativeSeismicDesignLevel
The methodology permits the user to select the seismic design level considered appropriate for
the study region and to define a mix of seismic design levels for each model building type. The
1599
HazusMHTechnicalManual
building damage functions provided are based on current-Code provisions and represent buildings
of modern design and construction. Most buildings in a study region will likely not be of modern
design and construction (i.e., do not conform to 1976 UBC, 1985 NEHRP Provisions, or later
editions of these model Codes). For many study regions, particularly those in the Central and
Eastern United States, seismic provisions may not be enforced (or only adopted very recently).
Building damage functions for new buildings designed and constructed to meet modern-Code
provisions should not be used for older, non-complying buildings.

The building damage functions represent specific cells of a three by three matrix that defines
three seismic design levels (High, Moderate and Low) and, for each of these design levels, three
seismic performance levels (Inferior, Ordinary and Superior), as shown in Table 5.19. For
completeness, cells representing Special buildings of Chapter 6 (Essential Facilities) are also
included in the matrix.

Table5.19SeismicDesignandPerformanceLevelsofDefaultBuildingDamageFunctions(and
ApproximateStructuralStrengthandDuctility)
SeismicPerformanceLevel
SeismicDesign
Level
Superior
1
Ordinary Inferior
High
(UBCZone4)
SpecialHighCode
MaximumStrength
MaximumDuctility
HighCode
HighStrength
HighDuctility

ModerateStrength
Mod./LowDuctility
Moderate
(UBCZone2B)
SpecialModerateCode
High/Mod.Strength
HighDuctility
ModerateCode
ModerateStrength
ModerateDuctility

LowStrength
LowDuctility
Low
(UBCZone1)
SpecialLowCode
Mod./LowStrength
ModerateDuctility
LowCode
LowStrength
LowDuctility
PreCode
MinimalStrength
MinimalDuctility
1. SeeChapter6forSpecialHighCode,ModerateCodeandLowCodebuildingdamagefunctions.

Table 5.19 also defines the approximate structural strength and ductility attributes of buildings
occupying each of the nine cells. The design level is defined by Seismic Zones of the Uniform
15100
Chapter5DirectPhysicalDamageGeneralBuildingStock
Building Code (UBC), since most buildings in the United States that have been designed for
earthquakes used some version of the UBC. Table 5.20 relates UBC seismic zones to seismic
design regions of the NEHRP Provisions.

Expert users may tailor the damage functions to their study area of interest by determining the
appropriate fraction of each building type that conforms essentially to modern-Code provisions
(based on age of construction). Buildings deemed not to conform to modern-Code provisions
should be assigned a lower seismic design level, or defined as Pre-Code buildings if not
seismically designed. For instance, older buildings located in High-Code seismic design areas
should be evaluated using damage functions for either Moderate-Code buildings or Pre-Code
buildings, for buildings that pre-date seismic codes. Table 5.20 provides guidance for selecting
appropriate building damage functions based on building location (i.e., seismic region) and
building age. The years shown as break-off points should be considered very approximate and
may not be appropriate for many seismic regions, particularly regions of low and moderate
seismicity where seismic codes have not been routinely enforced.

15101
HazusMHTechnicalManual
Table5.20GuidelinesforSelectionofDamageFunctionsforTypicalBuildingsBasedonUBC
SeismicZoneandBuildingAge
UBCSeismicZone
(NEHRPMapArea)
Post1975 1941 1975 Pre1941
Zone4
(MapArea7)
HighCode ModerateCode PreCode
(W1=ModerateCode)
Zone3
(MapArea6)
ModerateCode ModerateCode PreCode
(W1=ModerateCode)
Zone2B
(MapArea5)
ModerateCode LowCode PreCode
(W1=LowCode)
Zone2A
(MapArea4)
LowCode LowCode PreCode
(W1=LowCode)
Zone1
(MapArea2/3)
LowCode PreCode
(W1=LowCode)
PreCode
(W1=LowCode)
Zone0
(MapArea1)
PreCode
(W1=LowCode)
PreCode
(W1=LowCode)
PreCode
(W1=LowCode)

The guidelines given in Table 5.20 assume that buildings in the study region are not designed for
wind. The user should consider the possibility that mid-rise and high-rise buildings could be
designed for wind and may have considerable lateral strength (though not ductility), even if not
designed for earthquake. Users must be knowledgeable about the type and history of construction
in the study region of interest and apply engineering judgment in assigning the fraction of each
building type to a seismic design group.

5.7.2 DevelopmentofDamageFunctionsforOtherBuildings

For a building type other than one of the 36 described in Table 5.1, expert users should select a
set of building damage functions that best represents the type of construction, strength and
ductility of the building type of interest. Such buildings include rehabilitated structures that have
improved seismic capacity. For example, URM (Pre-Code) buildings retrofitted in accordance
with Division 88, the Los Angeles City Ordinance to reduce the risk of life loss, demonstrated
15102
Chapter5DirectPhysicalDamageGeneralBuildingStock
significantly improved seismic performance during the 1994 Northridge earthquake [SSC, 1995].
Structural damage to these buildings would be better estimated using either essential facility
damage functions of either Low-Code or Moderate-Code RM1 buildings.


5.8 BuildingDamageReferences
Allen and Hoshall, Jack R. Benjamin and Associates, and Systan Inc. 1985. An Assessment of
DamageandCasualtiesforSixCitiesintheCentralUnitedStatesResultingfromEarthquakesin
theNewMadridSeismicZone.PreparedforFEMA.
Applied Technology Council (ATC). 1985. ATC13 Earthquake Damage Evaluation Data for
California.RedwoodCity,California:AppliedTechnologyCouncil.
Central United States Earthquake Preparedness Project. 1990. Estimated Future Earthquake
LossesforSt.LouisCityandCounty,Missouri.FEMAReport192.Washington,D.C.
Culver, Charles G., H.S. Lew, Gary C. Hart, and Clarkson W. Pinkham. 1975. Natural Hazards
Evaluation of Existing Buildings. U.S. Department of Commerce Building Science Series 61.
Washington,D.C.
Czarnecki,RobertMartin,1973.EarthquakeDamagetoTallBuildings.StructuresPublicationNo.
359,Dept.ofCivilEngineering,M.I.T.
Earthquake Engineering Research Institute (EERI). 1994. Expected Seismic Performance of
Buildings.Oakland,California.
Federal Emergency Management Agency (FEMA). 1996a. NEHRP Guidelines for the Seismic
RehabilitationofBuildings.FEMA273.Washington,D.C.
FederalEmergencyManagementAgency(FEMA).1996b.NEHRPCommentaryontheGuidelines
fortheSeismicRehabilitationofBuildings.FEMA274.Washington,D.C.
Federal Emergency Management Agency (FEMA). 1997a. NEHRP Recommended Provisions for
SeismicRegulationsforNewBuildings.Part1Provisions.FEMA222A.Washington,D.C.
Federal Emergency Management Agency (FEMA). 1995b. NEHRP Recommended Provisions for
SeismicRegulationsforNewBuildings.Part2Commentary.FEMA223A.Washington,D.C.
Federal Emergency Management Agency (FEMA). 1992. NEHRP Handbook for the Seismic
EvaluationofExistingBuildings.FEMA178.Washington,D.C.
Ferritto,J.M.1980.AnEconomicAnalysisofSeismicUpgradingforExistingBuildings.NavalCivil
EngineeringLaboratory,TMNo.518017.PortHueneme,California.
Ferritto, J.M. 1982. An Economic Analysis of Earthquake Design Levels. Naval Civil Engineering
Laboratory,TNNo.N1640.PortHueneme,California.
Ferritto, J.M. 1983. An Economic Analysis of Earthquake Design Levels for New Concrete
Construction.NavalCivilEngineeringLaboratory,TNNo.N1671.PortHueneme,California.
15103
HazusMHTechnicalManual
Hasselman,T.K.,RonaldT.Eguchi,andJohnH.Wiggins.1980.AssessmentofDamageabilityfor
ExistingBuildingsinaNaturalHazardsEnvironment.VolumeI:Methodology.RedondoBeach,
California.
Kennedy, R. P., C. A. Cornell, R. L. Campbell, S. Kaplan, and H. F. Perla. 1980. Probabilistic
SeismicSafetyofanExistingNuclearPowerPlant.NuclearEngineeringandDesign59(2):315
38.
Kustu,O.,D.D.Miller,andS.T.Brokken.1982.DevelopmentofDamageFunctionsforHighRise
BuildingComponents.PreparedbyURS/JohnA.Blume&AssociatesfortheU.S.Departmentof
Energy.SanFrancisco,California
Kustu,O.,D.D.Miller,andR.E.Scholl.1983.AComputerizedMethodforPredictingEarthquake
Losses In Urban Areas Final Technical Report for Phase II. Prepared by URS/John A. Blume &
AssociatesfortheUSGS.SanFrancisco,California.
Lizundia, B., W. Dong, and W. T. Holmes. 1993. Analysis of Unreinforced Masonry Building
Damage Patterns in the Loma Prieta Earthquake and Improvement of Loss Estimation
Methodologies.Rutherford&Chekene,Inc.SanFrancisco,California.
Mahaney, James A., Terrence F. Paret, Bryan E. Kehoe, and Sigmund A. Freeman. 1993. The
Capacity Spectrum Method for Evaluating Structural Response during the Loma Prieta
Earthquake.Proceedingsofthe1993UnitedStatesNationalEarthquakeConference,Memphis,
Tennessee.Vol.2,Pages501510.
Newmark, N.M. and W.J. Hall. 1982. Earthquake Spectra and Design, Earthquake Engineering
ResearchInstituteMonograph.
OAKEngineeringInc.(OAK),"DevelopmentofDamageFunctionsforBuildings,"1994
Rihal, Satwant S. 1980. Racking Tests of Nonstructural Building Partitions. School of
ArchitectureandEnvironmentalDesign,CaliforniaPolytechnicStateUniversity,SanLuisObispo,
California.
State of California, Division of the State Architect (DSA). 1996. Earthquake Hazard Mitigation
TechnologyApplicationGuidelines,Sacramento,California.
State of California, Seismic Safety Commission (SSC). 1996. Seismic Evaluation and Retrofit of
ConcreteBuildings.ReportNo.SSC9601,Sacramento,California.
StateofCalifornia,SeismicSafetyCommission(SSC).1995.TurningLosstoGain.ReportNo.SSC
9501,Sacramento,California.
Steinbrugge, K.V., and S.T. Algermissen. 1990. Earthquake Losses to SingleFamily Dwellings:
CaliforniaExperience.USGSBulletin1939.UnitedStatesGeologicalSurvey.
Whitman, Robert V., SheuTien Hong and John W. Reed. 1973. Damage Statistics for HighRise
Buildings in the Vicinity of the San Fernando Earthquake. Structures Publication No. 363, Dept.
ofCivilEngineering,M.I.T.Boston,Massachusetts.
Whitman, Robert V. 1973. Damage Probability Matrices for Prototype Buildings. Structures
PublicationNo.380,Dept.ofCivilEngineering,M.I.T.Boston,Massachusetts.
15104
Chapter5DirectPhysicalDamageGeneralBuildingStock
Whitman, Robert V., Tarek S. Aziz and Earl H. Wong. 1977. Preliminary Correlations Between
Earthquake Damage and Strong Ground Motion. Structures Publication No. 564, Dept. of Civil
Engineering,M.I.T.Boston,Massachusetts.
Wong,EarlHom.1975.CorrelationsBetweenEarthquakeDamageandStrongGroundMotion.
Dept.ofCivilEngineering,M.I.T.,Boston,Massachusetts.

61
HazusMHTechnicalManual

Chapter 6
Direct Physical Damage - Essential and High Potential Loss Facilities


6.1 Introduction

This chapter describes methods for determining the probability of Slight, Moderate,
Extensive and Complete damage to essential facilities. These methods are identical to
those of Chapter 5 that describe damage to Code buildings, except that certain essential
facilities are represented by Special building damage functions. Special building
damage functions are appropriate for evaluation of essential facilities when the user
anticipates above-Code seismic performance for these facilities. The flowchart of the
methodology highlighting the essential and high potential loss facility damage
components and showing its relationship to other components is shown in Flowchart 6.1.

This chapter also provides guidance for high potential loss (HPL) facilities. The
methodology highlighting the Direct Physical Damage is shown in Flowchart 6.1.

6.1.1 Scope

The scope of this chapter includes: (1) classification of essential facilities, (2) building
damage functions for Special buildings, (3) methods for estimation of earthquake damage
to essential facilities, given knowledge of the model building type and seismic design
level, and an estimate of earthquake demand, and (3) guidance for expert users, including
estimation of damage to high potential loss (HPL) facilities.

Special buildings and their damage functions are described in Sections 6.2 through 6.5.
Evaluation of damage to essential facilities is given in Section 6.6 and guidance for
expert users is given in Section 6.7. Typically, sections of Chapter 6 reference (rather
than repeat) material of the corresponding section of Chapter 5.

6.1.2 Essential Facility Classification

Facilities that provide services to the community and those that should be functional
following an earthquake are considered to be essential facilities. Examples of essential
facilities include hospitals, police stations, fire stations, emergency operations centers
(EOCs) and schools. The methodology adopted for damage assessment of such facilities
is explained in this section.

Essential facilities are classified on the basis of facility function and, in the case of
hospitals, size. Table 6.1 lists the classes of essential facilities used in the Methodology.
Hospitals are classified on the basis of number of beds, since the structural and
nonstructural systems of a hospital are related to the size of the hospital (i.e., to the
number of beds it contains).
62
Chapter6DirectPhysicalDamageEssentialandHighPotentialLossFacilities

8. Lifelines-
Utility
Systems
4. Ground Motion 4. Ground Failure
Direct Physical
Damage
6. Essential and
High Potential
Loss Facilities
12. Debris 10. Fire 15. Economic 14. Shelter 9. Inundation 11. HazMat
16. Indirect
Economic
Losses
Potential Earth Science Hazards
Direct Economic/
Social Losses
Induced Physical
Damage
7. Lifelines-
Transportation
Systems
5. General
Building
Stock
13. Casualities


Flowchart 6.1: Essential and High Potential Loss Facility Component Relationship to
other Components in the Methodology

63
HazusMHTechnicalManual

Table 6.1 Classification of Essential Facilities

No. Label Occupancy Class Description
Medical Care Facilities
1 EFHS Small Hospital Hospital with less than 50 Beds
2 EFHM Medium Hospital Hospital with beds between 50 & 150
3 EFHL Large Hospital Hospital with greater than 150 Beds
4 EFMC Medical Clinics Clinics, Labs, Blood Banks
Emergency Response
5 EFFS Fire Station
6 EFPS Police Station
7 EFEO Emergency Operation Centers
Schools
8 EFS1 Schools Primary/ Secondary Schools (K-12)
9 EFS2 Colleges/Universities Community and State Colleges, State
and Private Universities


It is the responsibility of the user to identify each essential facility as either a Code
building or a Special building of a particular model building type and seismic design
level. This chapter provides building damage functions for Special buildings that have
significantly better than average seismic capacity. Chapter 5 provides building damage
functions for Code-buildings. If the user is not able to determine that the essential facility
is significantly better than average, then the facility should be modeled using Code
building damage functions (i.e., the same methods as those developed in Chapter 5 for
general building stock).

6.1.3 Input Requirements and Output Information

Input required to estimate essential facility damage using fragility and capacity curves
includes the following two items:

- model building type (including height) and seismic design level that represents the
essential facility (or type of essential facilities) of interest, and

- response spectrum (or PGA, for lifeline buildings, and PGD for ground failure
evaluation) at the essential facilitys site.

The response spectrum, PGA and PGD at the essential facility site are PESH outputs,
described in Chapter 4.

The output of fragility curves is an estimate of the cumulative probability of being in or
exceeding, each damage state for the given level of ground shaking (or ground failure).
Cumulative damage probabilities are differenced to create discrete damage state
probabilities, as described in Chapter 5 (Section 5.6). Discrete probabilities of damage
are used directly as inputs to induced physical damage and direct economic and social
loss modules, as shown in Flowchart 6.1.

64
Chapter6DirectPhysicalDamageEssentialandHighPotentialLossFacilities

Typically, the model building type (including height) is not known for each essential
facility and must be inferred from the inventory of essential facilities using the
occupancy/building type relationships described in Chapter 3. In general, performance of
essential facilities is not expected to be better than the typical building of the
representative model building type. Exceptions to this generalization include California
hospitals of recent (post-1973) construction.

6.1.4 Form of Damage Functions

Building damage functions for essential facilities are of the same form as those described
in Chapter 5 for general building stock. For each damage state, a lognormal fragility
curve relates the probability of damage to PGA, PGD or spectral demand determined by
the intersection of the model building types capacity curve and the demand spectrum.
Figure 6.1 provides an example of fragility curves for four damage states: Slight,
Moderate, Extensive and Complete.


Spectral Displacement (inches)
0.00
0.25
0.50
0.75
1.00
0.00 5.00 10.00 15.00 20.00 25.00
Slight Moderate Extensive Complete


Figure 6.1 Example Fragility Curves for Slight, Moderate, Extensive
and Complete Damage.


The fragility curves are driven by a PESH parameter. For ground failure, the PESH
parameter used to drive fragility curves is permanent ground displacement (PGD). For
ground shaking, the PESH parameter used to drive building fragility curves is peak
spectral response (either displacement or acceleration), or peak ground acceleration
(PGA) for essential lifeline facilities. Peak spectral response varies significantly for
65
HazusMHTechnicalManual

buildings that have different response properties and, therefore, requires knowledge of
these properties.

Building response is characterized by building capacity curves. These curves describe
the push-over displacement of each building type and seismic design level as a function
of laterally-applied earthquake load. Design-, yield- and ultimate-capacity points define
the shape of each building capacity curve. The Methodology estimates peak building
response as the intersection of the building capacity curve and the demand spectrum at
the buildings location.

The demand spectrum is the 5%-damped PESH input spectrum reduced for higher levels
of effective damping (e.g., effective damping includes both elastic damping and
hysteretic damping associated with post-yield cyclic response of the building). Figure 6.2
illustrates the intersection of a typical building capacity curve and a typical demand
spectrum (reduced for effective damping greater than 5% of critical).


Figure 6.2 Example Building Capacity Curve and Demand Spectrum.


6.2 Description of Model Building Types

The model building types used for essential facilities are identical to those used for
general building stock. These building types are described in Section 5.2 and listed in
Table 5.1. Typical nonstructural components of essential facilities include those
architectural, mechanical and electrical, and contents listed in Table 5.2 for general
building stock.

0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 1 2 3 4 5 6 7 8 9 10
Spectral Displacement (inches)
S
p
e
c
t
r
a
l

A
c
c
e
l
e
r
a
t
i
o
n

(
g
'
s
)
PESH Input Spectrum
(5% Damping)
Demand Spectrum
(Damping > 5%)
Capacity
Curve
Ultimate Capacity
Design Capacity
Yield Capacity
S
a
S
d
66
Chapter6DirectPhysicalDamageEssentialandHighPotentialLossFacilities

Essential facilities also include certain special equipment, such as emergency generators,
and certain special contents, such as those used to operate a hospital. Special equipment
and contents of essential facilities are considered to be acceleration-sensitive
nonstructural components of these facilities.

6.3 Description of Building Damage States

Building damage states for structural and nonstructural components of essential facilities
are the same as those described in Section 5.3 for general building stock.

6.4 Building Damage Due to Ground Shaking - Special Buildings

6.4.1 Overview

This section describes capacity and fragility curves used in the Methodology to estimate
the probability of Slight, Moderate, Extensive and Complete damage to Special buildings
of a given model building type designed to High-, Moderate-, or Low-Code seismic
standards. Special building damage functions are appropriate for evaluation of essential
facilities when the user anticipates above-Code seismic performance for these facilities.

Capacity curves and fragility curves for Special buildings of High-Code, Moderate-Code,
or Low-Code seismic design are based on modern code (e.g., 1976 Uniform Building
Code, 1996 NEHRP Provisions, or later editions of these model codes) design criteria for
various seismic design zones, as shown in Table 6.2. Additional description of seismic
design levels may be found in Section 6.7.

Table 6.2 Approximate Basis for Seismic Design Levels

Seismic Design Level
(I =1.5)
Seismic Zone
(1994 Uniform Building Code)
Map Area
(1994 NEHRP Provisions)
High-Code
Moderate-Code
Low-Code
4
2B
1
7
5
3


The capacity and fragility curves represent buildings designed and constructed to modern
seismic code provisions (e.g.,1994 UBC) using an importance factor of I = 1.5.
Moderate-Code and Low-Code seismic design levels are included for completeness.
Most essential facilities located in Seismic Zones O, T, 2A or 2B have not been designed
for Special building code criteria.
67
HazusMHTechnicalManual


6.4.2 Capacity Curves - Special Buildings

The building capacity curves for Special buildings are similar to those for the general
building stock (Chapter 5), but with increased strength. Each curve is described by three
control points that define model building capacity:

- Design Capacity
- Yield Capacity
- Ultimate Capacity

Design capacity represents the nominal building strength required by current model
seismic code provisions (e.g., 1994 UBC) including an importance factor of I =1.5.
Wind design is not considered in the estimation of design capacity and certain buildings
(e.g., taller buildings located in zones of low or moderate seismicity) may have a lateral
design strength considerably greater than based on seismic code provisions.

Yield capacity represents the true lateral strength of the building considering
redundancies in design, conservatism in code requirements and true (rather than nominal)
strength of materials. Ultimate capacity represents the maximum strength of the building
when the global structural system has reached a fully plastic state. An example building
capacity curve is shown in Figure 6-3.


Sa
Sd
Dy Du Dd
Au
Ay
Ad
Ultimate
Capacity
Yield
Capacity
Design
Capacity


Figure 6.3 Example Building Capacity Curve.

The building capacity curves for Special buildings are constructed based on the same
engineering properties (i.e., C
s
, T
e
, o
1
, o
2
, , , ) as those used to describe capacity
curves of Code buildings (i.e., Tables 5.4, 5.5 and 5.6), except for design strength, C
s
,
and ductility (). The design strength, C
s
, is approximately based on the lateral-force
design requirements of current seismic codes (e.g., 1994 NEHRP or 1994 UBC) using an
importance factor of I =1.5. Values of the ductility factor, , for Special buildings are
68
Chapter6DirectPhysicalDamageEssentialandHighPotentialLossFacilities

based on Code-building ductility increased by 1.33 for Moderate-Code buildings and by
1.2 for Low-Code buildings. The ductility parameter defines the displacement value of
capacity curve at the point where the curve reaches a fully plastic state.

Building capacity curves are assumed to have a range of possible properties that are
lognormally distributed as a function of the ultimate strength (A
u
) of each capacity curve.
Special building capacity curves represent median estimates of building capacity. The
variability of the capacity of each building type is assumed to be: |(A
u
) =0.15 for
Special buildings. An example construction of median, 84th percentile (+1|) and 16th
percentile (1|) building capacity curves for a typical building is illustrated in Figure 6.4.
Median capacity curves are intersected with demand spectra to estimate peak building
response. The variability of the capacity curves is used, with other sources of variability
and uncertainty, to define total fragility curve variability.

Figure 6.4 Example Construction of Median, +1| and -1| Building Capacity
Curves.

Tables 6.3a, 6.3b and 6.3c summarize yield capacity and ultimate capacity control points
for Special buildings of High-Code, Moderate-Code and Low-Code seismic design
levels, respectively. Note that for the following tables, shaded boxes indicate types that
are not permitted by current seismic codes.

69
HazusMHTechnicalManual


Table 6.3a Special Building Capacity Curves - High-Code
Seismic Design Level

Building Yield Capacity Point Ultimate Capacity Point
Type D
y
(in.) A
y
(g) D
u
(in.) A
u
(g)
W1 0.72 0.600 17.27 1.800
W2 0.94 0.600 18.79 1.500
S1L 0.92 0.375 22.00 1.124
S1M 2.66 0.234 42.60 0.702
S1H 6.99 0.147 83.83 0.440
S2L 0.94 0.600 15.03 1.200
S2M 3.64 0.500 38.82 1.000
S2H 11.62 0.381 92.95 0.762
S3 0.94 0.600 15.03 1.200
S4L 0.58 0.480 10.36 1.080
S4M 1.64 0.400 19.65 0.900
S4H 5.23 0.305 47.05 0.685
S5L
S5M
S5H
C1L 0.59 0.375 14.08 1.124
C1M 1.73 0.312 27.65 0.937
C1H 3.02 0.147 36.20 0.440
C2L 0.72 0.600 14.39 1.500
C2M 1.56 0.500 20.76 1.250
C2H 4.41 0.381 44.09 0.952
C3L
C3M
C3H
PC1 1.08 0.900 17.27 1.800
PC2L 0.72 0.600 11.51 1.200
PC2M 1.56 0.500 16.61 1.000
PC2H 4.41 0.381 35.27 0.762
RM1L 0.96 0.800 15.34 1.600
RM1M 2.08 0.667 22.14 1.333
RM2L 0.96 0.800 15.34 1.600
RM2M 2.08 0.667 22.14 1.333
RM2H 5.88 0.508 47.02 1.015
URML
URMM
MH 0.27 0.225 4.32 0.450


610
Chapter6DirectPhysicalDamageEssentialandHighPotentialLossFacilities

Table 6.3b Special Building Capacity Curves - Moderate-Code
Seismic Design Level

Building Yield Capacity Point Ultimate Capacity Point
Type D
y
(in.) A
y
(g) D
u
(in.) A
u
(g)
W1 0.54 0.450 12.95 1.350
W2 0.47 0.300 9.40 0.750
S1L 0.46 0.187 11.00 0.562
S1M 1.33 0.117 21.30 0.351
S1H 3.49 0.073 41.91 0.220
S2L 0.47 0.300 7.52 0.600
S2M 1.82 0.250 19.41 0.500
S2H 5.81 0.190 46.47 0.381
S3 0.47 0.300 7.52 0.600
S4L 0.29 0.240 5.18 0.540
S4M 0.82 0.200 9.83 0.450
S4H 2.61 0.152 23.53 0.343
S5L
S5M
S5H
C1L 0.29 0.187 7.04 0.562
C1M 0.86 0.156 13.83 0.468
C1H 1.51 0.073 18.10 0.220
C2L 0.36 0.300 7.19 0.750
C2M 0.78 0.250 10.38 0.625
C2H 2.21 0.190 22.05 0.476
C3L
C3M
C3H
PC1 0.54 0.450 8.63 0.900
PC2L 0.36 0.300 5.76 0.600
PC2M 0.78 0.250 8.31 0.500
PC2H 2.21 0.190 17.64 0.381
RM1L 0.48 0.400 7.67 0.800
RM1M 1.04 0.333 11.07 0.667
RM2L 0.48 0.400 7.67 0.800
RM2M 1.04 0.333 11.07 0.667
RM2H 2.94 0.254 23.51 0.508
URML
URMM
MH 0.27 0.225 4.32 0.450


611
HazusMHTechnicalManual

Table 6.3c Special Building Capacity Curves - Low-Code
Seismic Design Level

Building Yield Capacity Point Ultimate Capacity Point
Type D
y
(in.) A
y
(g) D
u
(in.) A
u
(g)
W1 0.36 0.300 6.48 0.900
W2 0.24 0.150 3.52 0.375
S1L 0.23 0.094 4.13 0.281
S1M 0.67 0.059 7.99 0.176
S1H 1.75 0.037 15.72 0.110
S2L 0.24 0.150 2.82 0.300
S2M 0.91 0.125 7.28 0.250
S2H 2.91 0.095 17.43 0.190
S3 0.24 0.150 2.82 0.300
S4L 0.14 0.120 1.94 0.270
S4M 0.41 0.100 3.69 0.225
S4H 1.31 0.076 8.82 0.171
S5L 0.18 0.150 2.16 0.300
S5M 0.51 0.125 4.09 0.250
S5H 1.63 0.095 9.80 0.190
C1L 0.15 0.094 2.64 0.281
C1M 0.43 0.078 5.19 0.234
C1H 0.75 0.037 6.79 0.110
C2L 0.18 0.150 2.70 0.375
C2M 0.39 0.125 3.89 0.313
C2H 1.10 0.095 8.27 0.238
C3L 0.18 0.150 2.43 0.338
C3M 0.39 0.125 3.50 0.281
C3H 1.10 0.095 7.44 0.214
PC1 0.27 0.225 3.24 0.450
PC2L 0.18 0.150 2.16 0.300
PC2M 0.39 0.125 3.11 0.250
PC2H 1.10 0.095 6.61 0.190
RM1L 0.24 0.200 2.88 0.400
RM1M 0.52 0.167 4.15 0.333
RM2L 0.24 0.200 2.88 0.400
RM2M 0.52 0.167 4.15 0.333
RM2H 1.47 0.127 8.82 0.254
URML 0.36 0.300 4.32 0.600
URMM 0.41 0.167 3.26 0.333
MH 0.27 0.225 4.32 0.450



612
Chapter6DirectPhysicalDamageEssentialandHighPotentialLossFacilities


6.4.3 Fragility Curves - Special Buildings

This section describes Special building fragility curves for Slight, Moderate, Extensive
and Complete structural damage states and Slight, Moderate, Extensive and Complete
nonstructural damage states. Each fragility curve is characterized by a median and a
lognormal standard deviation (|) value of PESH demand. Spectral displacement is the
PESH parameter used for structural damage and nonstructural damage to drift-sensitive
components. Spectral acceleration is the PESH parameter used for nonstructural damage
to acceleration-sensitive components.

Special building fragility curves for ground failure are the same as those of Code
buildings (Section 5.5).

6.4.3.1 Background

The form of the fragility curves for Special buildings is the same as that used for Code
buildings. The probability of being in, or exceeding, a given damage state is modeled as
a cumulative lognormal distribution. Given the appropriate PESH parameter (e.g.,
spectral displacement, S
d,
for structural damage), the probability of being in or exceeding
a damage state, ds, is modeled as follows:

| |
(

|
|
.
|

\
|
u =
ds d
ds
ds
,
d
d
S
S
ln
1
S P
|
(6-1)

where: S
d,ds
is the median value of spectral displacement at which the building
reaches the threshold of the damage state, ds,
|
ds
is the standard deviation of the natural logarithm of spectral
displacement of damage state, ds, and
u is the standard normal cumulative distribution function.

6.4.3.2 Structural Damage - Special Buildings

Structural damage states for Special buildings are based on drift ratios that are assumed to
be slightly higher than those of Code buildings of the same model building type and
seismic design level. It is difficult to quantify this improvement in displacement capacity
since it is a function not just of building type and design parameters, but also design
review and construction inspection. It is assumed that the improvement in displacement
capacity results in a 1.25 increase in drift capacity of each damage state for all Special
building types and seismic design levels. Special buildings perform better than Code
buildings due to increased structure strength (of the capacity curves) and increased
displacement capacity (of the fragility curves). In general, increased strength tends to
most improve building performance near yield and improved displacement capacity tends
to most improve the ultimate capacity of the building.

613
HazusMHTechnicalManual

Median values of Special building structural fragility are based on drift ratios (that
describe the threshold of damage states and the height of the building to point of push-
over mode displacement using the same approach as that of Code buildings (Section
5.4.3.2).

The variability of Special building structural damage is based on the same approach as
that of Code buildings (Section 5.4.3.3). The total variability of each structural damage
state, |
Sds
, is modeled by the combination of following three contributors to damage
variability:

- uncertainty in the damage state threshold of the structural system
(|
M(Sds)
=0.4, for all structural damage states and building types)

- variability in capacity (response) properties of the model building
type/seismic design level of interest (|
C(Au)
= 0.15 for Special
buildings), and

- variability in response due to the spatial variability of ground motion
demand (|
D(A)
=0.45 and |
C(V)
=0.50).is based on the dispersion
factor typical of the attenuation of large-magnitude earthquakes as in
the WUS (Chapter 4).

Each of these three contributors to damage state variability are assumed to be
lognormally distributed random variables. Capacity and demand are dependent
parameters and a convolution process is used to derive combined capacity/demand
variability of each structural damage state. Capacity/demand variability is then combined
with damage state uncertainty, as described in Section 5.4.3.3.

Tables 6.4a, 6.4b and 6.4c summarize median and lognormal standard deviation (|
Sds
)
values for Slight, Moderate, Extensive and Complete structural damage states of Special
buildings for High-Code, Moderate-Code and Low-Code seismic design levels,
respectively. Note that for the following tables, shaded boxes indicate types that are not
permitted by current seismic codes.
614
Chapter6DirectPhysicalDamageEssentialandHighPotentialLossFacilities

Table 6.4a Building Structural Fragility - High-Code Seismic Design Level

Table 6.4b Building Structural Fragility - Moderate-Code Seismic Design Level

BuildingProperties Interstory Drift at Spectral Displacement (inches)
Type Height (inches) Threshold of DamageState Slight Moderate Extensive Complete
Roof Modal Slight Moderate Extensive Complete Median Beta Median Beta Median Beta Median Beta
W1 168 126 0.0050 0.0124 0.0383 0.0937 0.63 0.76 1.56 0.77 4.82 0.78 11.81 0.96
W2 288 216 0.0050 0.0124 0.0383 0.0938 1.08 0.79 2.68 0.86 8.27 0.88 20.25 0.84
S1L 288 216 0.0075 0.0130 0.0294 0.0750 1.62 0.73 2.80 0.71 6.35 0.70 16.20 0.77
S1M 720 540 0.0050 0.0086 0.0196 0.0500 2.70 0.64 4.67 0.65 10.58 0.66 27.00 0.75
S1H 1872 1123 0.0037 0.0065 0.0147 0.0375 4.21 0.62 7.29 0.62 16.51 0.66 42.12 0.70
S2L 288 216 0.0063 0.0108 0.0292 0.0750 1.35 0.82 2.34 0.85 6.30 0.89 16.20 0.85
S2M 720 540 0.0042 0.0072 0.0194 0.0500 2.25 0.66 3.90 0.66 10.50 0.68 27.00 0.81
S2H 1872 1123 0.0031 0.0054 0.0146 0.0375 3.51 0.62 6.08 0.63 16.38 0.65 42.12 0.71
S3 180 135 0.0050 0.0087 0.0234 0.0656 0.68 0.77 1.17 0.81 3.16 0.89 8.86 0.89
S4L 288 216 0.0050 0.0087 0.0234 0.0656 1.08 0.88 1.87 0.92 5.05 0.98 14.18 0.87
S4M 720 540 0.0033 0.0058 0.0156 0.0437 1.80 0.70 3.12 0.67 8.41 0.70 23.62 0.90
S4H 1872 1123 0.0025 0.0043 0.0117 0.0328 2.81 0.66 4.87 0.66 13.13 0.70 36.86 0.81
S5L
S5M
S5H
C1L 240 180 0.0063 0.0108 0.0292 0.0750 1.13 0.80 1.95 0.82 5.25 0.84 13.50 0.81
C1M 600 450 0.0042 0.0072 0.0194 0.0500 1.87 0.66 3.25 0.67 8.75 0.66 22.50 0.84
C1H 1440 864 0.0031 0.0054 0.0146 0.0375 2.70 0.64 4.68 0.64 12.60 0.68 32.40 0.81
C2L 240 180 0.0050 0.0105 0.0289 0.0750 0.90 0.77 1.89 0.86 5.21 0.91 13.50 0.89
C2M 600 450 0.0033 0.0070 0.0193 0.0500 1.50 0.71 3.16 0.70 8.68 0.69 22.50 0.83
C2H 1440 864 0.0025 0.0053 0.0145 0.0375 2.16 0.64 4.55 0.65 12.51 0.66 32.40 0.79
C3L
C3M
C3H
PC1 180 135 0.0050 0.0087 0.0234 0.0656 0.68 0.79 1.17 0.81 3.16 0.86 8.86 1.00
PC2L 240 180 0.0050 0.0087 0.0234 0.0656 0.90 0.83 1.56 0.89 4.21 0.97 11.81 0.89
PC2M 600 450 0.0033 0.0058 0.0156 0.0438 1.50 0.76 2.60 0.74 7.01 0.73 19.69 0.88
PC2H 1440 864 0.0025 0.0043 0.0117 0.0328 2.16 0.65 3.75 0.66 10.10 0.70 28.35 0.81
RM1L 240 180 0.0050 0.0087 0.0234 0.0656 0.90 0.80 1.56 0.85 4.21 0.92 11.81 0.97
RM1M 600 450 0.0033 0.0058 0.0156 0.0438 1.50 0.73 2.60 0.75 7.01 0.75 19.69 0.80
RM2L 240 180 0.0050 0.0087 0.0234 0.0656 0.90 0.77 1.56 0.81 4.21 0.92 11.81 0.96
RM2M 600 450 0.0033 0.0058 0.0156 0.0438 1.50 0.72 2.60 0.72 7.01 0.72 19.69 0.77
RM2H 1440 864 0.0025 0.0043 0.0117 0.0328 2.16 0.63 3.75 0.65 10.10 0.66 28.35 0.76
URML
URMM
MH 120 120 0.0050 0.0100 0.0300 0.0875 0.60 0.81 1.20 0.89 3.60 0.97 10.50 0.86

Bui lding Properties I nterstory Drift at Spectral Displ acement(inches)
Type Hei ght(inches) Threshold of Damage State Sl ight Moderate Extensi ve Complete
Roof Modal Sli ght Moderate Extensive Compl ete Medi an Beta Median Beta Medi an Beta Median Beta
W1 168 126 0.0050 0.0150 0.0500 0.1250 0.63 0.66 1.89 0.72 6.30 0.72 15.75 0.91
W2 288 216 0.0050 0.0150 0.0500 0.1250 1.08 0.69 3.24 0.77 10.80 0.89 27.00 0.85
S1L 288 216 0.0075 0.0150 0.0375 0.1000 1.62 0.67 3.24 0.70 8.10 0.71 21.60 0.68
S1M 720 540 0.0050 0.0100 0.0250 0.0667 2.70 0.62 5.40 0.62 13.50 0.63 36.00 0.71
S1H 1872 1123 0.0037 0.0075 0.0188 0.0500 4.21 0.63 8.42 0.62 21.06 0.62 56.16 0.63
S2L 288 216 0.0063 0.0125 0.0375 0.1000 1.35 0.69 2.70 0.80 8.10 0.89 21.60 0.84
S2M 720 540 0.0042 0.0083 0.0250 0.0667 2.25 0.62 4.50 0.66 13.50 0.66 36.00 0.71
S2H 1872 1123 0.0031 0.0063 0.0188 0.0500 3.51 0.62 7.02 0.63 21.06 0.63 56.16 0.66
S3 180 135 0.0050 0.0100 0.0300 0.0875 0.68 0.66 1.35 0.71 4.05 0.80 11.81 0.90
S4L 288 216 0.0050 0.0100 0.0300 0.0875 1.08 0.77 2.16 0.82 6.48 0.92 18.90 0.91
S4M 720 540 0.0033 0.0067 0.0200 0.0583 1.80 0.69 3.60 0.67 10.80 0.68 31.50 0.82
S4H 1872 1123 0.0025 0.0050 0.0150 0.0438 2.81 0.62 5.62 0.63 16.85 0.65 49.14 0.73
S5L
S5M
S5H
C1L 240 180 0.0063 0.0125 0.0375 0.1000 1.13 0.69 2.25 0.74 6.75 0.82 18.00 0.81
C1M 600 450 0.0042 0.0083 0.0250 0.0667 1.87 0.63 3.75 0.65 11.25 0.66 30.00 0.71
C1H 1440 864 0.0031 0.0063 0.0188 0.0500 2.70 0.63 5.40 0.63 16.20 0.63 43.20 0.69
C2L 240 180 0.0050 0.0125 0.0375 0.1000 0.90 0.69 2.25 0.72 6.75 0.82 18.00 0.95
C2M 600 450 0.0033 0.0083 0.0250 0.0667 1.50 0.65 3.75 0.69 11.25 0.66 30.00 0.70
C2H 1440 864 0.0025 0.0063 0.0188 0.0500 2.16 0.62 5.40 0.63 16.20 0.64 43.20 0.69
C3L
C3M
C3H
PC1 180 135 0.0050 0.0100 0.0300 0.0875 0.68 0.63 1.35 0.74 4.05 0.79 11.81 0.96
PC2L 240 180 0.0050 0.0100 0.0300 0.0875 0.90 0.76 1.80 0.80 5.40 0.87 15.75 0.97
PC2M 600 450 0.0033 0.0067 0.0200 0.0583 1.50 0.66 3.00 0.73 9.00 0.72 26.25 0.73
PC2H 1440 864 0.0025 0.0050 0.0150 0.0438 2.16 0.62 4.32 0.64 12.96 0.65 37.80 0.74
RM1L 240 180 0.0050 0.0100 0.0300 0.0875 0.90 0.70 1.80 0.74 5.40 0.76 15.75 0.98
RM1M 600 450 0.0033 0.0067 0.0200 0.0583 1.50 0.63 3.00 0.68 9.00 0.70 26.25 0.70
RM2L 240 180 0.0050 0.0100 0.0300 0.0875 0.90 0.66 1.80 0.70 5.40 0.76 15.75 0.97
RM2M 600 450 0.0033 0.0067 0.0200 0.0583 1.50 0.63 3.00 0.70 9.00 0.69 26.25 0.68
RM2H 1440 864 0.0025 0.0050 0.0150 0.0438 2.16 0.63 4.32 0.63 12.96 0.63 37.80 0.65
URML
URMM
MH 120 120 0.0050 0.0100 0.0300 0.0875 0.60 0.81 1.20 0.89 3.60 0.97 10.50 0.86

615
HazusMHTechnicalManual

Table 6.4c Special Building Structural Fragility - Low-Code Seismic Design Level

6.4.3.3 Nonstructural Damage - Drift-Sensitive

Damage states of nonstructural drift-sensitive components of Special buildings are based
on the same drift ratios as those of Code buildings (Table 5.10). Even for essential
facilities, nonstructural components are typically not designed or detailed for special
earthquake displacements. Improvement in the performance of drift-sensitive
components of Special buildings is assumed to be entirely a function of drift reduction
due to the increased stiffness and strength of the structures of these buildings.

Median values of drift-sensitive nonstructural fragility curves are based on global
building displacement (in inches), calculated as the product of: (1) drift ratio, (2)
building height and (3) the fraction of building height at the location of push-over mode
displacement (o
2
).

The total variability of each nonstructural drift-sensitive damage state, |
NSDds
, is modeled
by the combination of following three contributors to damage variability:

- uncertainty in the damage state threshold of nonstructural components
(|
M(NSDds)
=0.5, for all structural damage states and building types,

- variability in capacity (response) properties of the model building type
that contains the nonstructural components of interest (|
C(Au)
=0.15
for Special buildings, and

Building Properties Interstory Drift at Spectral Displacement (inches)
Type Height (inches) Threshold of DamageState Slight Moderate Extensive Complete
Roof Modal Slight Moderate Extensive Complete Median Beta Median Beta Median Beta Median Beta
W1 168 126 0.0050 0.0124 0.0383 0.0937 0.63 0.80 1.56 0.81 4.82 0.88 11.81 1.01
W2 288 216 0.0050 0.0124 0.0383 0.0938 1.08 0.89 2.68 0.89 8.27 0.86 20.25 0.97
S1L 288 216 0.0075 0.0119 0.0253 0.0625 1.62 0.73 2.58 0.73 5.47 0.75 13.50 0.93
S1M 720 540 0.0050 0.0080 0.0169 0.0417 2.70 0.66 4.30 0.70 9.12 0.78 22.50 0.91
S1H 1872 1123 0.0037 0.0060 0.0127 0.0313 4.21 0.64 6.72 0.66 14.23 0.68 35.10 0.86
S2L 288 216 0.0063 0.0100 0.0250 0.0625 1.35 0.89 2.16 0.89 5.40 0.88 13.50 0.97
S2M 720 540 0.0042 0.0067 0.0167 0.0417 2.25 0.67 3.60 0.68 9.00 0.74 22.50 0.92
S2H 1872 1123 0.0031 0.0050 0.0125 0.0313 3.51 0.62 5.62 0.63 14.04 0.68 35.10 0.84
S3 180 135 0.0050 0.0080 0.0201 0.0547 0.68 0.89 1.08 0.90 2.71 0.98 7.38 0.85
S4L 288 216 0.0050 0.0080 0.0200 0.0547 1.08 0.98 1.73 0.95 4.33 0.97 11.81 0.98
S4M 720 540 0.0033 0.0053 0.0134 0.0364 1.80 0.69 2.88 0.72 7.22 0.81 19.68 0.98
S4H 1872 1123 0.0025 0.0040 0.0100 0.0273 2.81 0.66 4.50 0.67 11.26 0.78 30.71 0.93
S5L 288 216 0.0038 0.0075 0.0188 0.0438 0.81 1.00 1.62 1.00 4.05 1.03 9.45 0.91
S5M 720 540 0.0025 0.0050 0.0125 0.0292 1.35 0.74 2.70 0.72 6.75 0.78 15.75 0.94
S5H 1872 1123 0.0019 0.0037 0.0094 0.0219 2.11 0.67 4.21 0.69 10.53 0.74 24.57 0.90
C1L 240 180 0.0063 0.0100 0.0250 0.0625 1.13 0.85 1.80 0.85 4.50 0.88 11.25 0.95
C1M 600 450 0.0042 0.0067 0.0167 0.0417 1.87 0.70 3.00 0.69 7.50 0.75 18.75 0.95
C1H 1440 864 0.0031 0.0050 0.0125 0.0313 2.70 0.66 4.32 0.71 10.80 0.79 27.00 0.95
C2L 240 180 0.0050 0.0096 0.0247 0.0625 0.90 0.91 1.72 0.94 4.44 1.01 11.25 0.90
C2M 600 450 0.0033 0.0064 0.0164 0.0417 1.50 0.76 2.86 0.74 7.40 0.74 18.75 0.94
C2H 1440 864 0.0025 0.0048 0.0123 0.0313 2.16 0.66 4.12 0.67 10.66 0.74 27.00 0.91
C3L 240 180 0.0038 0.0075 0.0188 0.0438 0.68 0.92 1.35 0.99 3.38 1.04 7.88 0.88
C3M 600 450 0.0025 0.0050 0.0125 0.0292 1.12 0.77 2.25 0.79 5.62 0.78 13.12 0.93
C3H 1440 864 0.0019 0.0038 0.0094 0.0219 1.62 0.68 3.24 0.69 8.10 0.70 18.90 0.88
PC1 180 135 0.0050 0.0080 0.0201 0.0547 0.68 0.89 1.08 0.95 2.71 1.00 7.38 0.96
PC2L 240 180 0.0050 0.0080 0.0201 0.0547 0.90 0.98 1.44 0.98 3.61 1.02 9.84 0.91
PC2M 600 450 0.0033 0.0053 0.0134 0.0364 1.50 0.76 2.40 0.75 6.02 0.75 16.40 0.94
PC2H 1440 864 0.0025 0.0040 0.0100 0.0273 2.16 0.66 3.46 0.68 8.66 0.73 23.63 0.92
RM1L 240 180 0.0050 0.0080 0.0201 0.0547 0.90 0.97 1.44 1.01 3.61 1.07 9.84 0.88
RM1M 600 450 0.0033 0.0053 0.0134 0.0364 1.50 0.78 2.40 0.78 6.02 0.78 16.40 0.94
RM2L 240 180 0.0050 0.0080 0.0201 0.0547 0.90 0.94 1.44 0.98 3.61 1.05 9.84 0.89
RM2M 600 450 0.0033 0.0053 0.0134 0.0364 1.50 0.76 2.40 0.75 6.02 0.75 16.40 0.92
RM2H 1440 864 0.0025 0.0040 0.0100 0.0273 2.16 0.66 3.46 0.67 8.66 0.80 23.63 0.89
URML 180 135 0.0038 0.0075 0.0187 0.0438 0.51 0.89 1.01 0.91 2.53 0.96 5.91 1.09
URMM 420 315 0.0025 0.0050 0.0125 0.0292 0.79 0.81 1.57 0.84 3.94 0.87 9.19 0.82
MH 120 120 0.0050 0.0100 0.0300 0.0875 0.60 0.81 1.20 0.89 3.60 0.97 10.50 0.86

616
Chapter6DirectPhysicalDamageEssentialandHighPotentialLossFacilities

- variability in response of the model building type due to the spatial
variability of ground motion demand (|
D(A)
=0.45 and |
C(V)
=0.50).

Each of these three contributors to damage state variability are assumed to be
lognormally distributed random variables. Capacity and demand are dependent
parameters and a convolution process is used to derive combined capacity/demand
variability of each nonstructural damage state. Capacity/demand variability is then
combined with damage state uncertainty, as described in Section 5.4.3.3.

Tables 6.5a, 6.5b and 6.5c summarize median and lognormal standard deviation (|
NSDds
)
values for Slight, Moderate, Extensive and Complete damage states of nonstructural drift-
sensitive components of Special buildings for High-Code, Moderate-Code and Low-Code
seismic design levels, respectively. Note that for the following tables, shaded boxes
indicate types that are not permitted by current seismic codes.

617
HazusMHTechnicalManual


Table 6.5a Special Building Nonstructural Drift-Sensitive Fragility - High-Code
Seismic Design Level

Building Median Spectral Displacement (inches) and Logstandard Deviation (Beta)
Type Slight Moderate Extensive Complete
Median Beta Median Beta Median Beta Median Beta
W1 0.50 0.74 1.01 0.77 3.15 0.79 6.30 0.78
W2 0.86 0.76 1.73 0.77 5.40 0.88 10.80 0.93
S1L 0.86 0.72 1.73 0.76 5.40 0.75 10.80 0.74
S1M 2.16 0.68 4.32 0.68 13.50 0.70 27.00 0.73
S1H 4.49 0.70 8.99 0.69 28.08 0.69 56.16 0.70
S2L 0.86 0.74 1.73 0.77 5.40 0.90 10.80 0.95
S2M 2.16 0.70 4.32 0.72 13.50 0.73 27.00 0.72
S2H 4.49 0.71 8.99 0.69 28.08 0.70 56.16 0.73
S3 0.54 0.70 1.08 0.76 3.38 0.83 6.75 0.93
S4L 0.86 0.81 1.73 0.84 5.40 0.93 10.80 1.00
S4M 2.16 0.76 4.32 0.74 13.50 0.75 27.00 0.82
S4H 4.49 0.70 8.99 0.71 28.08 0.72 56.16 0.80
S5L
S5M
S5H
C1L 0.72 0.77 1.44 0.76 4.50 0.84 9.00 0.88
C1M 1.80 0.71 3.60 0.71 11.25 0.72 22.50 0.71
C1H 3.46 0.70 6.91 0.69 21.60 0.71 43.20 0.75
C2L 0.72 0.76 1.44 0.76 4.50 0.80 9.00 0.94
C2M 1.80 0.74 3.60 0.76 11.25 0.73 22.50 0.74
C2H 3.46 0.69 6.91 0.69 21.60 0.71 43.20 0.75
C3L
C3M
C3H
PC1 0.54 0.69 1.08 0.78 3.38 0.85 6.75 0.88
PC2L 0.72 0.80 1.44 0.83 4.50 0.90 9.00 1.03
PC2M 1.80 0.75 3.60 0.80 11.25 0.77 22.50 0.77
PC2H 3.46 0.70 6.91 0.71 21.60 0.73 43.20 0.82
RM1L 0.72 0.74 1.44 0.80 4.50 0.80 9.00 0.94
RM1M 1.80 0.70 3.60 0.77 11.25 0.77 22.50 0.77
RM2L 0.72 0.74 1.44 0.76 4.50 0.78 9.00 0.96
RM2M 1.80 0.71 3.60 0.78 11.25 0.74 22.50 0.74
RM2H 3.46 0.69 6.91 0.69 21.60 0.71 43.20 0.74
URML
URMM
MH 0.48 0.85 0.96 0.92 3.00 0.98 6.00 0.99

618
Chapter6DirectPhysicalDamageEssentialandHighPotentialLossFacilities

Table 6.5b Special Building Nonstructural Drift-Sensitive Fragility - Moderate-
Code Seismic Design Level

Building Median Spectral Displacement (inches) and Logstandard Deviation (Beta)
Type Slight Moderate Extensive Complete
Median Beta Median Beta Median Beta Median Beta
W1 0.50 0.77 1.01 0.82 3.15 0.84 6.30 0.87
W2 0.86 0.84 1.73 0.88 5.40 0.93 10.80 0.93
S1L 0.86 0.78 1.73 0.78 5.40 0.78 10.80 0.76
S1M 2.16 0.71 4.32 0.71 13.50 0.73 27.00 0.81
S1H 4.49 0.69 8.99 0.69 28.08 0.72 56.16 0.82
S2L 0.86 0.81 1.73 0.91 5.40 0.96 10.80 0.89
S2M 2.16 0.73 4.32 0.74 13.50 0.73 27.00 0.87
S2H 4.49 0.69 8.99 0.70 28.08 0.74 56.16 0.84
S3 0.54 0.82 1.08 0.86 3.38 0.97 6.75 0.95
S4L 0.86 0.89 1.73 0.97 5.40 1.02 10.80 0.94
S4M 2.16 0.76 4.32 0.74 13.50 0.84 27.00 0.97
S4H 4.49 0.71 8.99 0.73 28.08 0.83 56.16 0.94
S5L
S5M
S5H
C1L 0.72 0.80 1.44 0.86 4.50 0.88 9.00 0.88
C1M 1.80 0.73 3.60 0.72 11.25 0.74 22.50 0.89
C1H 3.46 0.71 6.91 0.71 21.60 0.79 43.20 0.93
C2L 0.72 0.84 1.44 0.87 4.50 0.95 9.00 1.00
C2M 1.80 0.79 3.60 0.76 11.25 0.76 22.50 0.88
C2H 3.46 0.70 6.91 0.71 21.60 0.77 43.20 0.87
C3L
C3M
C3H
PC1 0.54 0.82 1.08 0.87 3.38 0.93 6.75 1.02
PC2L 0.72 0.88 1.44 0.95 4.50 1.03 9.00 0.99
PC2M 1.80 0.84 3.60 0.77 11.25 0.79 22.50 0.95
PC2H 3.46 0.72 6.91 0.74 21.60 0.84 43.20 0.94
RM1L 0.72 0.86 1.44 0.88 4.50 0.99 9.00 1.04
RM1M 1.80 0.80 3.60 0.79 11.25 0.79 22.50 0.88
RM2L 0.72 0.81 1.44 0.86 4.50 0.97 9.00 1.03
RM2M 1.80 0.78 3.60 0.77 11.25 0.77 22.50 0.88
RM2H 3.46 0.71 6.91 0.71 21.60 0.74 43.20 0.87
URML
URMM
MH 0.48 0.85 0.96 0.92 3.00 0.98 6.00 0.99


619
HazusMHTechnicalManual

Table 6.5c Special Building Nonstructural Drift-Sensitive Fragility - Low-Code
Seismic Design Level

Building Median Spectral Displacement (inches) and Logstandard Deviation (Beta)
Type Slight Moderate Extensive Complete
Median Beta Median Beta Median Beta Median Beta
W1 0.50 0.83 1.01 0.86 3.15 0.88 6.30 1.00
W2 0.86 0.93 1.73 0.94 5.40 0.99 10.80 0.93
S1L 0.86 0.81 1.73 0.80 5.40 0.80 10.80 0.94
S1M 2.16 0.73 4.32 0.76 13.50 0.86 27.00 0.98
S1H 4.49 0.71 8.99 0.74 28.08 0.87 56.16 0.98
S2L 0.86 0.94 1.73 0.93 5.40 0.93 10.80 0.98
S2M 2.16 0.73 4.32 0.76 13.50 0.91 27.00 0.99
S2H 4.49 0.71 8.99 0.74 28.08 0.85 56.16 0.96
S3 0.54 0.89 1.08 0.96 3.38 1.01 6.75 0.90
S4L 0.86 1.02 1.73 0.99 5.40 0.95 10.80 1.01
S4M 2.16 0.76 4.32 0.84 13.50 0.95 27.00 1.04
S4H 4.49 0.74 8.99 0.87 28.08 0.96 56.16 1.03
S5L 0.86 1.04 1.73 1.04 5.40 1.00 10.80 0.99
S5M 2.16 0.78 4.32 0.84 13.50 0.97 27.00 1.04
S5H 4.49 0.76 8.99 0.87 28.08 0.96 56.16 1.03
C1L 0.72 0.90 1.44 0.92 4.50 0.93 9.00 0.93
C1M 1.80 0.74 3.60 0.77 11.25 0.94 22.50 1.00
C1H 3.46 0.75 6.91 0.86 21.60 0.97 43.20 1.03
C2L 0.72 0.93 1.44 0.99 4.50 1.06 9.00 0.92
C2M 1.80 0.80 3.60 0.80 11.25 0.91 22.50 1.00
C2H 3.46 0.73 6.91 0.80 21.60 0.93 43.20 1.01
C3L 0.72 0.99 1.44 1.05 4.50 1.06 9.00 0.93
C3M 1.80 0.84 3.60 0.83 11.25 0.95 22.50 1.01
C3H 3.46 0.76 6.91 0.84 21.60 0.96 43.20 1.03
PC1 0.54 0.92 1.08 0.99 3.38 1.07 6.75 1.02
PC2L 0.72 0.99 1.44 1.02 4.50 1.02 9.00 0.95
PC2M 1.80 0.81 3.60 0.82 11.25 0.95 22.50 1.02
PC2H 3.46 0.74 6.91 0.86 21.60 0.96 43.20 1.02
RM1L 0.72 0.98 1.44 1.06 4.50 1.08 9.00 0.94
RM1M 1.80 0.83 3.60 0.84 11.25 0.91 22.50 0.99
RM2L 0.72 0.94 1.44 1.03 4.50 1.07 9.00 0.92
RM2M 1.80 0.81 3.60 0.80 11.25 0.91 22.50 0.99
RM2H 3.46 0.74 6.91 0.79 21.60 0.92 43.20 1.01
URML 0.54 0.93 1.08 0.98 3.38 1.05 6.75 1.11
URMM 1.26 0.89 2.52 0.88 7.88 0.87 15.75 0.99
MH 0.48 0.85 0.96 0.92 3.00 0.98 6.00 0.99

620
Chapter6DirectPhysicalDamageEssentialandHighPotentialLossFacilities

6.4.3.4 Nonstructural Damage - Acceleration-Sensitive Components

Damage states of nonstructural acceleration-sensitive components of Special buildings
are based on the peak floor accelerations of Code buildings of seismic design level (Table
5.12) increased by a factor of 1.5. A factor of 1.5 on damage-state acceleration reflects
increased anchorage strength of nonstructural acceleration-sensitive components of
Special buildings.

The floor acceleration values are used directly as median values, assuming average
upper-floor demand is represented by response at the point of the push-over mode
displacement.

The total variability of each damage state, |
NSAds
, is modeled by the combination of
following three contributors to nonstructural acceleration-sensitive damage variability:

- uncertainty in the damage state threshold of nonstructural components
(|
M(NSAds)
=0.6, for all structural damage states and building types,

- variability in capacity (response) properties of the model building type
that contains the nonstructural components of interest (|
C(Au)
=0.15
for Special buildings, and

- variability in response of the model building type due to the spatial
variability of ground motion demand (|
D(A)
=0.45 and |
C(V)
=0.50).

Each of these three contributors to damage state variability are assumed to be
lognormally distributed random variables. Capacity and demand are dependent
parameters and a convolution process is used to derive combined capacity/demand
variability of each nonstructural damage state. Capacity/demand variability is then
combined with damage state uncertainty, as described in Section 5.4.3.3.

Tables 6.6a, 6.6b and 6.6c summarize median and lognormal standard deviation (|
NSAds
)
values for Slight, Moderate, Extensive and Complete damage states of nonstructural
acceleration-sensitive components of Special buildings for High-Code, Moderate-Code
and Low-Code seismic design levels, respectively.


621
HazusMHTechnicalManual

Table 6.6a Special Building Nonstructural Acceleration-Sensitive Fragility - High-
Code Seismic Design Level

Building Median Spectral Acceleration (g) and Logstandard Deviation (Beta)
Type Slight Moderate Extensive Complete
Median Beta Median Beta Median Beta Median Beta
W1 0.45 0.72 0.90 0.68 1.80 0.68 3.60 0.68
W2 0.45 0.69 0.90 0.67 1.80 0.68 3.60 0.68
S1L 0.45 0.66 0.90 0.67 1.80 0.67 3.60 0.67
S1M 0.45 0.66 0.90 0.67 1.80 0.68 3.60 0.68
S1H 0.45 0.67 0.90 0.66 1.80 0.66 3.60 0.66
S2L 0.45 0.66 0.90 0.67 1.80 0.66 3.60 0.66
S2M 0.45 0.68 0.90 0.65 1.80 0.65 3.60 0.65
S2H 0.45 0.67 0.90 0.65 1.80 0.65 3.60 0.65
S3 0.45 0.68 0.90 0.67 1.80 0.66 3.60 0.66
S4L 0.45 0.67 0.90 0.67 1.80 0.67 3.60 0.67
S4M 0.45 0.66 0.90 0.65 1.80 0.66 3.60 0.66
S4H 0.45 0.66 0.90 0.65 1.80 0.63 3.60 0.63
S5L
S5M
S5H
C1L 0.45 0.67 0.90 0.68 1.80 0.67 3.60 0.67
C1M 0.45 0.66 0.90 0.66 1.80 0.66 3.60 0.66
C1H 0.45 0.67 0.90 0.65 1.80 0.65 3.60 0.65
C2L 0.45 0.68 0.90 0.67 1.80 0.67 3.60 0.63
C2M 0.45 0.68 0.90 0.65 1.80 0.64 3.60 0.64
C2H 0.45 0.68 0.90 0.65 1.80 0.64 3.60 0.64
C3L
C3M
C3H
PC1 0.45 0.72 0.90 0.66 1.80 0.67 3.60 0.63
PC2L 0.45 0.68 0.90 0.67 1.80 0.66 3.60 0.66
PC2M 0.45 0.67 0.90 0.64 1.80 0.65 3.60 0.65
PC2H 0.45 0.66 0.90 0.64 1.80 0.63 3.60 0.63
RM1L 0.45 0.73 0.90 0.66 1.80 0.68 3.60 0.64
RM1M 0.45 0.69 0.90 0.65 1.80 0.64 3.60 0.64
RM2L 0.45 0.71 0.90 0.66 1.80 0.67 3.60 0.63
RM2M 0.45 0.70 0.90 0.65 1.80 0.64 3.60 0.64
RM2H 0.45 0.69 0.90 0.65 1.80 0.64 3.60 0.64
URML
URMM
MH 0.38 0.66 0.75 0.67 1.50 0.67 3.00 0.67


622
Chapter6DirectPhysicalDamageEssentialandHighPotentialLossFacilities

Table 6.6b Special Building Nonstructural Acceleration-Sensitive Fragility -
Moderate-Code Seismic Design Level

Building Median Spectral Acceleration (g) and Logstandard Deviation (Beta)
Type Slight Moderate Extensive Complete
Median Beta Median Beta Median Beta Median Beta
W1 0.38 0.71 0.75 0.68 1.50 0.68 3.00 0.65
W2 0.38 0.67 0.75 0.68 1.50 0.68 3.00 0.68
S1L 0.38 0.67 0.75 0.67 1.50 0.68 3.00 0.68
S1M 0.38 0.67 0.75 0.67 1.50 0.67 3.00 0.67
S1H 0.38 0.67 0.75 0.66 1.50 0.66 3.00 0.66
S2L 0.38 0.66 0.75 0.66 1.50 0.68 3.00 0.68
S2M 0.38 0.65 0.75 0.65 1.50 0.64 3.00 0.64
S2H 0.38 0.65 0.75 0.65 1.50 0.65 3.00 0.65
S3 0.38 0.66 0.75 0.66 1.50 0.66 3.00 0.66
S4L 0.38 0.67 0.75 0.66 1.50 0.65 3.00 0.65
S4M 0.38 0.65 0.75 0.65 1.50 0.65 3.00 0.65
S4H 0.38 0.65 0.75 0.65 1.50 0.65 3.00 0.65
S5L
S5M
S5H
C1L 0.38 0.68 0.75 0.66 1.50 0.68 3.00 0.68
C1M 0.38 0.66 0.75 0.65 1.50 0.65 3.00 0.65
C1H 0.38 0.65 0.75 0.65 1.50 0.65 3.00 0.65
C2L 0.38 0.67 0.75 0.67 1.50 0.67 3.00 0.67
C2M 0.38 0.65 0.75 0.64 1.50 0.66 3.00 0.66
C2H 0.38 0.65 0.75 0.64 1.50 0.64 3.00 0.64
C3L
C3M
C3H
PC1 0.38 0.67 0.75 0.67 1.50 0.65 3.00 0.65
PC2L 0.38 0.66 0.75 0.66 1.50 0.64 3.00 0.64
PC2M 0.38 0.64 0.75 0.64 1.50 0.64 3.00 0.64
PC2H 0.38 0.64 0.75 0.65 1.50 0.65 3.00 0.65
RM1L 0.38 0.67 0.75 0.67 1.50 0.67 3.00 0.67
RM1M 0.38 0.65 0.75 0.64 1.50 0.66 3.00 0.66
RM2L 0.38 0.67 0.75 0.67 1.50 0.67 3.00 0.67
RM2M 0.38 0.65 0.75 0.64 1.50 0.66 3.00 0.66
RM2H 0.38 0.65 0.75 0.64 1.50 0.64 3.00 0.64
URML
URMM
MH 0.38 0.66 0.75 0.67 1.50 0.67 3.00 0.67


623
HazusMHTechnicalManual

Table 6.6c Special Building Nonstructural Acceleration-Sensitive Fragility - Low-
Code Seismic Design Level

Building Median Spectral Acceleration (g) and Logstandard Deviation (Beta)
Type Slight Moderate Extensive Complete
Median Beta Median Beta Median Beta Median Beta
W1 0.30 0.71 0.60 0.68 1.20 0.66 2.40 0.65
W2 0.30 0.66 0.60 0.66 1.20 0.69 2.40 0.69
S1L 0.30 0.66 0.60 0.68 1.20 0.68 2.40 0.68
S1M 0.30 0.66 0.60 0.68 1.20 0.68 2.40 0.68
S1H 0.30 0.67 0.60 0.67 1.20 0.67 2.40 0.67
S2L 0.30 0.65 0.60 0.68 1.20 0.68 2.40 0.68
S2M 0.30 0.65 0.60 0.67 1.20 0.67 2.40 0.67
S2H 0.30 0.64 0.60 0.67 1.20 0.67 2.40 0.67
S3 0.30 0.65 0.60 0.67 1.20 0.67 2.40 0.67
S4L 0.30 0.65 0.60 0.68 1.20 0.68 2.40 0.68
S4M 0.30 0.64 0.60 0.68 1.20 0.68 2.40 0.68
S4H 0.30 0.64 0.60 0.67 1.20 0.67 2.40 0.67
S5L 0.30 0.65 0.60 0.68 1.20 0.68 2.40 0.68
S5M 0.30 0.64 0.60 0.67 1.20 0.67 2.40 0.67
S5H 0.30 0.64 0.60 0.67 1.20 0.67 2.40 0.67
C1L 0.30 0.65 0.60 0.68 1.20 0.68 2.40 0.68
C1M 0.30 0.64 0.60 0.67 1.20 0.67 2.40 0.67
C1H 0.30 0.67 0.60 0.67 1.20 0.67 2.40 0.67
C2L 0.30 0.66 0.60 0.66 1.20 0.65 2.40 0.65
C2M 0.30 0.63 0.60 0.65 1.20 0.65 2.40 0.65
C2H 0.30 0.63 0.60 0.66 1.20 0.66 2.40 0.66
C3L 0.30 0.65 0.60 0.67 1.20 0.67 2.40 0.67
C3M 0.30 0.63 0.60 0.66 1.20 0.66 2.40 0.66
C3H 0.30 0.63 0.60 0.67 1.20 0.67 2.40 0.67
PC1 0.30 0.66 0.60 0.65 1.20 0.65 2.40 0.65
PC2L 0.30 0.65 0.60 0.68 1.20 0.68 2.40 0.68
PC2M 0.30 0.63 0.60 0.67 1.20 0.67 2.40 0.67
PC2H 0.30 0.64 0.60 0.66 1.20 0.66 2.40 0.66
RM1L 0.30 0.66 0.60 0.66 1.20 0.65 2.40 0.65
RM1M 0.30 0.64 0.60 0.65 1.20 0.65 2.40 0.65
RM2L 0.30 0.66 0.60 0.66 1.20 0.66 2.40 0.66
RM2M 0.30 0.64 0.60 0.65 1.20 0.65 2.40 0.65
RM2H 0.30 0.63 0.60 0.65 1.20 0.65 2.40 0.65
URML 0.30 0.68 0.60 0.66 1.20 0.64 2.40 0.64
URMM 0.30 0.64 0.60 0.65 1.20 0.65 2.40 0.65
MH 0.38 0.66 0.75 0.67 1.50 0.67 3.00 0.67


624
Chapter6DirectPhysicalDamageEssentialandHighPotentialLossFacilities



6.4.4 Structural Fragility Curves - Equivalent Peak Ground Acceleration

Structural damage fragility curves are expressed in terms of an equivalent value of PGA
(rather than spectral displacement) for evaluation of Special buildings that are
components of lifelines. Only structural damage functions are developed based on PGA,
since structural damage is considered the most appropriate measure of damage for lifeline
facilities. Similar methods could be used to develop nonstructural damage functions
based on PGA. In this case, capacity curves are not necessary to estimate building
response and PGA is used directly as the PESH input to building fragility curves.

This section provides equivalent-PGA fragility curves for Special buildings based on the
structural damage functions of Tables 6.4a - 6.4c and standard spectrum shape properties
of Chapter 4. These functions have the same format and are based on the same approach
and assumptions as those described in Section 5.4.4 for development equivalent-PGA
fragility curves for Code buildings.

The values given in Tables 6.7a through 6.7c are appropriate for use in the evaluation of
scenario earthquakes whose demand spectrum shape is based on, or similar to, large-
magnitude, WUS ground shaking at soil sites (reference spectrum shape). For evaluation
of building damage due to scenario earthquakes whose spectra are not similar to the
reference spectrum shape, damage-state median parameters may be adjusted to better
represent equivalent-PGA structural fragility for the spectrum shape of interest.

Median values of equivalent-PGA are adjusted for: (1) the site condition (if different
from Site Class D) and (2) the ratio of long-period spectral response (i.e., S
A1
) to PGA (if
different from a value of 1.5, the ratio of S
A1
to PGA of the reference spectrum shape).
Damage-state variability is not adjusted assuming that the variability associated with
ground shaking (although different for different source/site conditions) when combined
with the uncertainty in damage-state threshold, is approximately the same for all demand
spectrum shapes.

Equivalent-PGA medians, given in Tables 6.7a through 6.7c for the reference spectrum
shape, are adjusted to represent other spectrum shapes using the spectrum shape ratios of
Tables 5.14 and 5.15, the soil amplification factor, F
V
, and Equation (5-6). In general,
implementation of Equation (5-6) requires information on earthquake magnitude and
source-to-site distance to estimate the spectrum shape ratio for rock sites, and 1-second
period spectral acceleration at the site (to estimate the soil amplification factor). Note
that for the following tables, shaded boxes indicate types that are not permitted by current
seismic codes.


625
HazusMHTechnicalManual

Table 6.7a Equivalent-PGA Structural Fragility - Special High-Code
Seismic Design Level

Building Median Equivalent-PGA (g) and Logstandard Deviation (Beta)
Type Slight Moderate Extensive Complete
Median Beta Median Beta Median Beta Median Beta
W1 0.32 0.64 0.78 0.64 2.00 0.64 3.22 0.64
W2 0.35 0.64 0.82 0.64 1.76 0.64 3.13 0.64
S1L 0.25 0.64 0.44 0.64 0.92 0.64 2.17 0.64
S1M 0.17 0.64 0.34 0.64 0.85 0.64 2.10 0.64
S1H 0.13 0.64 0.26 0.64 0.65 0.64 1.73 0.64
S2L 0.33 0.64 0.58 0.64 1.10 0.64 2.07 0.64
S2M 0.18 0.64 0.35 0.64 0.97 0.64 2.34 0.64
S2H 0.14 0.64 0.27 0.64 0.81 0.64 2.13 0.64
S3 0.19 0.64 0.36 0.64 0.79 0.64 1.44 0.64
S4L 0.34 0.64 0.54 0.64 1.04 0.64 1.91 0.64
S4M 0.21 0.64 0.37 0.64 0.98 0.64 2.27 0.64
S4H 0.16 0.64 0.32 0.64 0.90 0.64 2.29 0.64
S5L
S5M
S5H
C1L 0.29 0.64 0.51 0.64 1.07 0.64 2.06 0.64
C1M 0.19 0.64 0.36 0.64 1.02 0.64 2.48 0.64
C1H 0.14 0.64 0.28 0.64 0.83 0.64 2.03 0.64
C2L 0.33 0.64 0.66 0.64 1.42 0.64 2.40 0.64
C2M 0.22 0.64 0.49 0.64 1.24 0.64 2.97 0.64
C2H 0.15 0.64 0.37 0.64 1.11 0.64 2.80 0.64
C3L
C3M
C3H
PC1 0.25 0.64 0.48 0.64 1.02 0.64 1.86 0.64
PC2L 0.32 0.64 0.51 0.64 1.03 0.64 1.78 0.64
PC2M 0.22 0.64 0.40 0.64 0.92 0.64 2.25 0.64
PC2H 0.15 0.64 0.30 0.64 0.83 0.64 2.13 0.64
RM1L 0.39 0.64 0.65 0.64 1.52 0.64 2.53 0.64
RM1M 0.25 0.64 0.50 0.64 1.15 0.64 2.76 0.64
RM2L 0.34 0.64 0.59 0.64 1.41 0.64 2.36 0.64
RM2M 0.22 0.64 0.43 0.64 1.05 0.64 2.65 0.64
RM2H 0.15 0.64 0.30 0.64 0.89 0.64 2.58 0.64
URML
URMM
MH 0.16 0.64 0.26 0.64 0.45 0.64 0.88 0.64


626
Chapter6DirectPhysicalDamageEssentialandHighPotentialLossFacilities

Table 6.7b Equivalent-PGA Structural Fragility - Special Moderate-Code
Seismic Design Level

Building Median Equivalent-PGA (g) and Logstandard Deviation (Beta)
Type Slight Moderate Extensive Complete
Median Beta Median Beta Median Beta Median Beta
W1 0.32 0.64 0.59 0.64 1.32 0.64 2.08 0.64
W2 0.28 0.64 0.51 0.64 1.00 0.64 1.83 0.64
S1L 0.20 0.64 0.31 0.64 0.60 0.64 1.29 0.64
S1M 0.16 0.64 0.28 0.64 0.60 0.64 1.27 0.64
S1H 0.13 0.64 0.22 0.64 0.51 0.64 1.17 0.64
S2L 0.27 0.64 0.37 0.64 0.67 0.64 1.27 0.64
S2M 0.17 0.64 0.28 0.64 0.69 0.64 1.40 0.64
S2H 0.14 0.64 0.23 0.64 0.63 0.64 1.44 0.64
S3 0.18 0.64 0.26 0.64 0.46 0.64 0.86 0.64
S4L 0.26 0.64 0.36 0.64 0.61 0.64 1.17 0.64
S4M 0.18 0.64 0.29 0.64 0.69 0.64 1.33 0.64
S4H 0.16 0.64 0.26 0.64 0.66 0.64 1.42 0.64
S5L
S5M
S5H
C1L 0.23 0.64 0.33 0.64 0.63 0.64 1.22 0.64
C1M 0.17 0.64 0.28 0.64 0.70 0.64 1.38 0.64
C1H 0.14 0.64 0.23 0.64 0.59 0.64 1.15 0.64
C2L 0.26 0.64 0.44 0.64 0.77 0.64 1.34 0.64
C2M 0.20 0.64 0.35 0.64 0.81 0.64 1.63 0.64
C2H 0.15 0.64 0.30 0.64 0.78 0.64 1.63 0.64
C3L
C3M
C3H
PC1 0.24 0.64 0.33 0.64 0.63 0.64 1.05 0.64
PC2L 0.24 0.64 0.35 0.64 0.59 0.64 1.06 0.64
PC2M 0.19 0.64 0.29 0.64 0.62 0.64 1.27 0.64
PC2H 0.15 0.64 0.25 0.64 0.60 0.64 1.30 0.64
RM1L 0.31 0.64 0.44 0.64 0.79 0.64 1.33 0.64
RM1M 0.24 0.64 0.36 0.64 0.74 0.64 1.65 0.64
RM2L 0.28 0.64 0.41 0.64 0.74 0.64 1.27 0.64
RM2M 0.21 0.64 0.32 0.64 0.69 0.64 1.58 0.64
RM2H 0.15 0.64 0.25 0.64 0.64 0.64 1.53 0.64
URML
URMM
MH 0.16 0.64 0.26 0.64 0.45 0.64 0.88 0.64

627
HazusMHTechnicalManual

Table 6.7c Equivalent-PGA Structural Fragility - Special Low-Code
Seismic Design Level

Building Median Equivalent-PGA (g) and Logstandard Deviation (Beta)
Type Slight Moderate Extensive Complete
Median Beta Median Beta Median Beta Median Beta
W1 0.28 0.64 0.50 0.64 1.00 0.64 1.51 0.64
W2 0.21 0.64 0.34 0.64 0.68 0.64 1.10 0.64
S1L 0.16 0.64 0.23 0.64 0.42 0.64 0.71 0.64
S1M 0.15 0.64 0.23 0.64 0.42 0.64 0.73 0.64
S1H 0.13 0.64 0.20 0.64 0.40 0.64 0.71 0.64
S2L 0.19 0.64 0.25 0.64 0.44 0.64 0.74 0.64
S2M 0.16 0.64 0.24 0.64 0.52 0.64 0.88 0.64
S2H 0.14 0.64 0.21 0.64 0.50 0.64 0.93 0.64
S3 0.14 0.64 0.18 0.64 0.30 0.64 0.57 0.64
S4L 0.19 0.64 0.23 0.64 0.38 0.64 0.68 0.64
S4M 0.16 0.64 0.23 0.64 0.47 0.64 0.81 0.64
S4H 0.15 0.64 0.23 0.64 0.48 0.64 0.87 0.64
S5L 0.18 0.64 0.26 0.64 0.41 0.64 0.68 0.64
S5M 0.14 0.64 0.24 0.64 0.50 0.64 0.80 0.64
S5H 0.13 0.64 0.24 0.64 0.51 0.64 0.84 0.64
C1L 0.17 0.64 0.22 0.64 0.39 0.64 0.67 0.64
C1M 0.15 0.64 0.23 0.64 0.48 0.64 0.80 0.64
C1H 0.13 0.64 0.20 0.64 0.39 0.64 0.66 0.64
C2L 0.19 0.64 0.27 0.64 0.44 0.64 0.79 0.64
C2M 0.16 0.64 0.26 0.64 0.56 0.64 0.93 0.64
C2H 0.14 0.64 0.25 0.64 0.56 0.64 0.96 0.64
C3L 0.17 0.64 0.25 0.64 0.39 0.64 0.65 0.64
C3M 0.14 0.64 0.23 0.64 0.46 0.64 0.75 0.64
C3H 0.12 0.64 0.22 0.64 0.48 0.64 0.79 0.64
PC1 0.18 0.64 0.24 0.64 0.38 0.64 0.65 0.64
PC2L 0.18 0.64 0.23 0.64 0.36 0.64 0.66 0.64
PC2M 0.16 0.64 0.22 0.64 0.45 0.64 0.79 0.64
PC2H 0.14 0.64 0.21 0.64 0.45 0.64 0.81 0.64
RM1L 0.22 0.64 0.29 0.64 0.44 0.64 0.80 0.64
RM1M 0.19 0.64 0.26 0.64 0.50 0.64 0.92 0.64
RM2L 0.20 0.64 0.27 0.64 0.41 0.64 0.77 0.64
RM2M 0.17 0.64 0.24 0.64 0.47 0.64 0.88 0.64
RM2H 0.14 0.64 0.22 0.64 0.49 0.64 0.92 0.64
URML 0.19 0.64 0.28 0.64 0.47 0.64 0.68 0.64
URMM 0.14 0.64 0.22 0.64 0.38 0.64 0.70 0.64
MH 0.16 0.64 0.26 0.64 0.45 0.64 0.88 0.64


628
Chapter6DirectPhysicalDamageEssentialandHighPotentialLossFacilities

6.5 Damage Due to Ground Failure - Special Buildings

Damage to Special buildings due to ground failure is assumed to be the same as the
damage to Code buildings for the same amount of permanent ground deformation (PGD).
Fragility curves developed in Section 5.5 for Code buildings are also appropriate for
prediction of damage to Special buildings due to ground failure.

6.6 Evaluation of Building Damage - Essential Facilities

6.6.1 Overview

Special building capacity and fragility curves for structural and nonstructural systems are
used to predict essential facility damage when the user is able to determine that the
essential facility is superior to a typical building of the model building type and design
level of interest. If such a determination cannot be made by the user, then the Code
building functions of Chapter 5 are used to evaluate essential building damage. These
criteria are summarized in Table 6.8.

Table 6.8 Criteria for Evaluating Essential Facility Damage

Evaluate Essential Facility Using: User Deems Essential Facility to be:
Code building damage functions
(High-Code, Moderate-Code, Low-
Code and Pre-Code functions of
Chapter 5)
Typical of the model building type and
seismic design level of interest (i.e., no
special seismic protection of components)
Special building damage functions
(High-Code, Moderate-Code and
Low-Code functions of Chapter 6)
Superior to the model building type and
seismic design level of interest (e.g., 50
percent stronger lateral-force-resisting
structural system, and special anchorage
and bracing of nonstructural components)


During an earthquake, the essential facilities may be damaged either by ground shaking,
ground failure, or both. Essential facilities are evaluated separately for the two modes of,
ground shaking and ground failure, and the resulting damage-state probabilities combined
for evaluation of loss.

6.6.2 Damage Due to Ground Shaking

Damage to essential facilities due to ground shaking uses the same methods as those
described in Section 5.6.2 for Code buildings, with the exception that Special buildings
are assumed to have less degradation and greater effective damping than Code buildings.

629
HazusMHTechnicalManual

6.6.2.1 Demand Spectrum Reduction for Effective Damping - Special Buildings

Demand spectra for evaluation of damage to Special buildings are constructed using the
same approach, assumptions and formulas as those described in Section 5.6.2.1 for Code
buildings, except values of the degradation factor, k, that defines the effective amount of
hysteretic damping as a function of duration are different for Special buildings.
Degradation factors for Special buildings are given in Table 6.9.

Figure 6.5 shows typical demand spectra (spectral acceleration plotted as a function of
spectral displacement) for three demand levels. These three demand levels represent
Short (k =0.90), Moderate (k =0.60) and Long (k =0.40) duration ground shaking,
respectively. Also shown in the figure is the building capacity curve of a low-rise Special
building (Moderate-Code seismic design) that was used to estimate effective damping.
The intersection of the capacity curve with each of the three demand spectra illustrates
the significance of duration (damping) on building response.
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 1 2 3 4 5 6 7 8 9 10
Spectral Displacement (inches)
S
p
e
c
t
r
a
l

A
c
c
e
l
e
r
a
t
i
o
n

(
g
'
s
)
PESH Spectrum - 5% Damping
Demand Spectrum - Long Duration
Demand Spectrum - Moderate Duration
Demand Spectrum - Short Duration
Building Capacity Curve
0.3 sec.
1.0 sec.

Figure 6.5 Example Demand Spectra - Special Building
(M = 7.0 at 20 km, WUS, Site Class E).

Comparison of Figure 6.5 with Figure 5.7 (same example building and PESH demand,
except capacity curve and damping represents Code building properties) illustrates the
significance of increased strength and damping (reduced degradation) of Special
buildings on the reduction of building displacement. In this case, the Special building
displaces only about one-half as much as a comparable Code building for the same level
of PESH demand. Forces on nonstructural acceleration-sensitive components are not
reduced, but are slightly increased due to the higher strength of the Special building.

630
Chapter6DirectPhysicalDamageEssentialandHighPotentialLossFacilities

Table 6.9 Special Building Degradation Factor (k) as a Function of Short,
Moderate and Long Earthquake Duration

Building Type High-Code Design Moderate-Code Design Low-Code Design
No. Label Short
Moderate
Long Short
Moderate
Long Short
Moderate
Long
1 W1 1.0 1.0 0.7 1.0 0.8 0.5 0.9 0.6 0.3
2 W2 1.0 0.8 0.6 0.9 0.6 0.4 0.8 0.4 0.2
3 S1L 1.0 0.8 0.6 0.9 0.6 0.4 0.8 0.4 0.2
4 S1M 1.0 0.8 0.6 0.9 0.6 0.4 0.8 0.4 0.2
5 S1H 1.0 0.8 0.6 0.9 0.6 0.4 0.8 0.4 0.2
6 S2L 0.8 0.6 0.4 0.7 0.5 0.3 0.6 0.4 0.2
7 S2M 0.8 0.6 0.4 0.7 0.5 0.3 0.6 0.4 0.2
8 S2H 0.8 0.6 0.4 0.7 0.5 0.3 0.6 0.4 0.2
9 S3 0.8 0.6 0.4 0.7 0.5 0.3 0.6 0.4 0.2
10 S4L 0.8 0.6 0.4 0.7 0.5 0.3 0.6 0.4 0.2
11 S4M 0.8 0.6 0.4 0.7 0.5 0.3 0.6 0.4 0.2
12 S4H 0.8 0.6 0.4 0.7 0.5 0.3 0.6 0.4 0.2
13 S5L 0.6 0.4 0.2 0.6 0.4 0.2 0.6 0.4 0.2
14 S5M 0.6 0.4 0.2 0.6 0.4 0.2 0.6 0.4 0.2
15 S5H 0.6 0.4 0.2 0.6 0.4 0.2 0.6 0.4 0.2
16 C1L 1.0 0.8 0.6 0.9 0.6 0.4 0.8 0.4 0.2
17 C1M 1.0 0.8 0.6 0.9 0.6 0.4 0.8 0.4 0.2
18 C1H 1.0 0.8 0.6 0.9 0.6 0.4 0.8 0.4 0.2
19 C2L 1.0 0.8 0.6 0.9 0.6 0.4 0.8 0.4 0.2
20 C2M 1.0 0.8 0.6 0.9 0.6 0.4 0.8 0.4 0.2
21 C2H 1.0 0.8 0.6 0.9 0.6 0.4 0.8 0.4 0.2
22 C3L 0.6 0.4 0.2 0.6 0.4 0.2 0.6 0.4 0.2
23 C3M 0.6 0.4 0.2 0.6 0.4 0.2 0.6 0.4 0.2
24 C3H 0.6 0.4 0.2 0.6 0.4 0.2 0.6 0.4 0.2
25 PC1 0.8 0.6 0.4 0.7 0.5 0.3 0.6 0.4 0.2
26 PC2L 0.8 0.6 0.4 0.7 0.5 0.3 0.6 0.4 0.2
27 PC2M 0.8 0.6 0.4 0.7 0.5 0.3 0.6 0.4 0.2
28 PC2H 0.8 0.6 0.4 0.7 0.5 0.3 0.6 0.4 0.2
29 RM1L 1.0 0.8 0.6 0.9 0.6 0.4 0.8 0.4 0.2
30 RM1M 1.0 0.8 0.6 0.9 0.6 0.4 0.8 0.4 0.2
31 RM2L 1.0 0.8 0.6 0.9 0.6 0.4 0.8 0.4 0.2
32 RM2M 1.0 0.8 0.6 0.9 0.6 0.4 0.8 0.4 0.2
33 RM2H 1.0 0.8 0.6 0.9 0.6 0.4 0.8 0.4 0.2
34 URML 0.6 0.4 0.2 0.6 0.4 0.2 0.6 0.4 0.2
35 URMM 0.6 0.4 0.2 0.6 0.4 0.2 0.6 0.4 0.2
36 MH 0.9 0.6 0.4 0.9 0.6 0.4 0.9 0.6 0.4


6.6.2.2 Damage State Probability

Structural and nonstructural fragility curves of essential facilities are evaluated for
spectral displacement and spectral acceleration defined by the intersection of the capacity
and demand curves. Each of these curves describe the cumulative probability of being in
or exceeding, a particular damage state. Nonstructural components (both drift- and
acceleration-sensitive components) may, in some cases, be dependent on the structural
damage state (e.g., Complete structural damage may cause Complete nonstructural
damage). The Methodology assumes nonstructural damage states to be independent of
631
HazusMHTechnicalManual

structural damage states. Cumulative probabilities are differenced to obtain discrete
probabilities of being in each of the five damage states. This process is shown
schematically in Figure 6.6.


Figure 6.6 Example Essential Facility Damage Estimation Process.



6.6.3 Combined Damage Due to Ground Failure and Ground Shaking

Damage to essential facilities is based either on Code building damage functions or
Special building damage functions. Code building damage due to ground shaking is
combined with damage due to ground shaking as specified in Section 5.6.3. Special
building damage due to ground failure (Section 6.5.2) is combined with damage due to
ground shaking (Section 6.6.2.2) using the same approach, assumptions and formulas as
those given in Section 5.6.3 for Code buildings.

6.6.4 Combined Damage to Essential Facilities

S
C
C
S S
C
M
E
M
E
M
E
632
Chapter6DirectPhysicalDamageEssentialandHighPotentialLossFacilities

Combined ground shaking/ground failure damage to the model building type and design
level of interest (either a Special or a Code building) represents combined damage to the
essential facility.

6.7 Guidance for Expert Users

This section provides guidance for users who are seismic/structural experts interested in
modifying essential facility damage functions supplied with the methodology. This
section also provides the expert user with guidance regarding the selection of the
appropriate mix of design levels for the region of interest.

6.7.1 Selection of Representative Seismic Design Level

The methodology permits the user to select the seismic design level considered
appropriate for each essential facility and to designate the facility as a Special building,
when designed and constructed to above-Code standards. In general, performance of
essential facilities is not expected to be better than the typical (Code) building of the
representative model building type. Exceptions to this generalization include California
hospitals of recent (post-1973) construction. If the user is not able to determine that the
essential facility is significantly better than average, then the facility should be modeled
using Code building damage functions (i.e., same methods as those developed in Chapter
5 for general building stock).

Table 6.10 provides guidance for selecting appropriate building damage functions for
essential facilities based on design vintage. These guidelines are applicable to the
following facilities:

1. hospitals and other medical facilities having surgery or emergency treatment
areas,
2. fire and police stations, and
3. municipal government disaster operation and communication centers deemed (for
design) to be vital in emergencies,

provided that seismic codes (e.g., Uniform Building Code) were adopted and enforced in
the study area of interest. Such adoption and enforcement is generally true for
jurisdictions of California, but may not be true other areas.


633
HazusMHTechnicalManual

Table 6.10 Guidelines for Selection of Damage Functions for Essential Facilities
Based on UBC Seismic Zone and Building Age

UBC Seismic Zone
(NEHRP Map Area)
Post-1973 1941 - 1973 Pre-1941
Zone 4
(Map Area 7)
Special High-Code Moderate-Code Pre-Code
(W1 =Moderate-Code)
Zone 3
(Map Area 6)
Special Moderate-Code Moderate-Code Pre-Code
(W1 =Moderate-Code)
Zone 2B
(Map Area 5)
Moderate-Code Low-Code Pre-Code
(W1 =Low-Code)
Zone 2A
(Map Area 4)
Low-Code Low-Code Pre-Code
(W1 =Low-Code)
Zone 1
Map Area 2/3)
Low-Code Pre-Code
(W1 =Low-Code)
Pre-Code
(W1 =Low-Code)
Zone 0
(Map Area 1)
Pre-Code
(W1 =Low-Code)
Pre-Code
(W1 =Low-Code)
Pre-Code
(W1 =Low-Code)

The guidelines given in Table 6.1 assume that essential buildings in the study region are
not designed for wind. The user should consider the possibility that mid-rise and high-
rise facilities could be designed for wind and may have considerable lateral strength, even
if not designed for earthquake. Users must be knowledgeable about the type and history
of construction in the study region of interest and apply engineering judgment in
assigning essential facilities to a building type and seismic design level.

6.7.2 High Potential Loss Facilities

6.7.2.1 Introduction

This section describes damage evaluation of high potential loss (HPL) facilities. HPL
facilities are likely to cause heavy earthquake losses, if significantly damaged. Examples
of such facilities include nuclear power plants, certain military and industrial facilities,
dams, etc.

6.7.2.2 Input Requirements and Output Information

The importance of these facilities (in terms of potential earthquake losses) suggests that
damage assessment be done in a special way as compared to ordinary buildings. Each
HPL facility should be treated on an individual basis by users who have sufficient
expertise to evaluate damage to such facilities. Required input to the damage evaluation
module includes the following items:

- capacity curves that represents median (typical) properties of the HPL facility
structure, or a related set of engineering parameters, such as period, yield strength,
and ultimate capacity, that may be used by seismic/structural engineering experts
with the methods of Chapter 5 to select representative damage functions,

634
Chapter6DirectPhysicalDamageEssentialandHighPotentialLossFacilities

- fragility curves for the HPL facility under consideration, or related set engineering
parameters, that can be used by seismic/structural engineering experts with the
methods of Chapter 5 to select appropriate damage functions.

The direct output (damage estimate) from implementation of the fragility curves is an
estimate of the probability of being in, or exceeding, each damage state for the given
level of ground shaking. This output is used directly as an input to other damage or loss
estimation methods or combined with inventory information to predict the distribution of
damage as a function of facility type, and geographical location. In the latter case, the
number and geographical location of facilities of interest would be a required input to the
damage estimation method.

6.7.2.3 Form of Damage Functions and Damage Evaluation
The form of user-supplied HPL facility damage functions should be the same as that of
buildings (Chapter 5) and their use in the methodology would be similar to that of
essential facilities.

6.8 Essential Facility and HPL Damage References
Refer to Section 5.8 for building damage references.

6.9 Restoration Curves
Restoration curves are based on generic ATC-13 data for the social function
classifications of interest and are approximated as normal curves characterized by a mean
and a standard deviation. The parameters of these restoration curves are given in Table
6.11 and are fully user-editable

Table 6.11 Generic Restoration Functions for Essential Facilities

EF
Class
Description Slight
Mean
Slight
Sigma
Moderat
e Mean
Moderat
e Sigma
Extensiv
e Mean
Extensiv
e Sigma
Complet
e Mean
Complet
e Sigma
EDFLT
Default for
Emergency
Response Facility
5 1 20 2 90 10 180 20
EFEO
Emergency
Operation Centers
5 1 20 2 90 10 180 20
EFFS Fire Station 5 1 20 2 90 10 180 20
EFHL
Large Hospital
(greater than 150
beds)
5 1 20 2 90 10 180 20
EFHM
Medium Hospital
(50 to 150 Beds)
5 1 20 2 90 10 180 20
EFHS
Small Hospital
(less than 50
Beds)
5 1 20 2 90 10 180 20
EFMC
Medical Clinics
and Labs
5 1 20 2 90 10 180 20
EFPS Police Station 5 1 20 2 90 10 180 20
EFS1
Grade Schools
(Primary and High
Schools)
5 1 20 2 90 10 180 20
EFS2
Colleges/Universiti
es
5 1 20 2 90 10 180 20
FDFLT
Default for Fire
Station
5 1 20 2 90 10 180 20
MDFL Default for Medical 5 1 20 2 90 10 180 20
635
HazusMHTechnicalManual

T
PDFLT Default for Police 5 1 20 2 90 10 180 20
SDFLT Default for School 5 1 20 2 90 10 180 20

71
HazusMHTechnicalManual
Chapter 7
Direct Physical Damage to Lifelines - Transportation Systems

This chapter describes the methodology for estimating direct physical damage to
Transportation Systems, which include the following seven systems:

- Highway
- Railway
- Light Rail
- Bus
- Port
- Ferry
- Airport

The flowchart of the overall methodology, highlighting the transportation system module
and its relationship to other modules, is shown in Flowchart 7.1.

7.1 Highway Transportation System

7.1.1 Introduction

This section presents an earthquake loss estimation methodology for a highway
transportation system. This system consists of roadways, bridges and tunnels. Roads
located on soft soil or fill or which cross a surface fault rupture can experience failure
resulting in loss of functionality. Bridges that fail usually result in significant disruption
to the transportation network, especially bridges that cross waterways. Likewise, tunnels
are often not redundant, and major disruption to the transportation system is likely to
occur should a tunnel become non-functional. Past earthquake damage reveals that
bridges and tunnels are vulnerable to both ground shaking and ground failure, while roads
are significantly affected by ground failure alone.

7.1.2 Scope

The scope of this section includes development of methods for estimation of earthquake
damage to a highway transportation system given knowledge of the system's components
(i.e., roadways, bridges, or tunnels), the classification of each component (e.g., for
roadways, whether the road is a major road or urban road), and the ground motion (i.e.
peak ground acceleration and/or permanent ground deformation).

Damage states describing the level of damage to each highway system component are
defined (i.e. slight/minor, moderate, extensive or complete). Damage states are related to
a damage ratio defined as the ratio of repair to replacement cost for evaluation of direct
economic loss. Component restoration curves are provided for each damage state to
evaluate loss of function. Restoration curves describe the fraction or percentage of the
component that is expected to be open or operational as a function of time following the
72
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
earthquake. For example, an extensively damaged roadway link might be closed (0%
functional) immediately following the earthquake, but 100% functional after 30 days.


8. Lifelines-
Utility
Systems
4. Ground Motion 4. Ground Failure
Direct Physical
Damage
6. Essential and
High Potential
Loss Facilities
12. Debris 10. Fire 15. Economic 14. Shelter 9. Inundation 11. HazMat
16. Indirect
Economic
Losses
Potential Earth Science Hazards
Direct Economic/
Social Losses
Induced Physical
Damage
7. Lifelines-
Transportation
Systems
5. General
Building
Stock
13. Casualities



Flowchart 7.1 Transportation System Damage Relationship to Other Modules of the
Earthquake Loss Estimation Methodology
73
HazusMHTechnicalManual
Fragility curves are developed for each type of highway system component. These
curves describe the probability of reaching or exceeding each damage state given the
level of ground motion.

7.1.3 Input Requirements and Output Information

Descriptions of required input to estimate damages to each highway system are given
below.

Roadways

- Geographical location of roadway links (longitude and latitude of end nodes)
- Permanent ground deformation (PGD) at roadway link
- Roadway classification

Bridges

- Geographical location of bridge [longitude and latitude]
- Bridge classification
- Spectral accelerations at 0.3 sec and 1.0 sec, and PGD at bridge
- Peak Ground Acceleration (for PGD-related computations)

Tunnels

- Geographical location of tunnels [longitude and latitude]
- PGA and PGD at tunnel
- Tunnel Classification

Direct damage output for highway systems includes probability estimates of (1)
component functionality and (2) physical damage expressed in terms of the component's
damage ratio. Note that damage ratios, which are input to direct economic loss methods,
are described in Chapter 15.

Component functionality is described by the probability of damage state (immediately
following the earthquake) and by the associated fraction or percentage of the component
that is expected to be functional after a specified period of time. For example, a roadway
link might be found to have a 0.50 probability of extensive damage and on this basis
would have a 0.50 probability that the road would be: (1) closed immediately, (2)
partially open after a 3-day restoration period and (3) fully open after a 1-month
restoration period.

Interdependence of components on overall system functionality is not addressed by the
methodology. Such considerations require a network system analysis that would be
performed separately by a highway system expert.

74
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
7.1.4 Form of Damage Functions

Damage functions or fragility curves for all three highway system components mentioned
above are modeled as lognormally-distributed functions that give the probability of
reaching or exceeding different damage states for a given level of ground motion or
ground failure. Each fragility curve is characterized by a median value of ground motion
or ground failure and an associated dispersion factor (lognormal standard deviation).
Ground motion is quantified in terms of peak ground acceleration (PGA) and spectral
acceleration (Sa), and ground failure is quantified in terms of permanent ground
displacement (PGD).

- For roadways, fragility curves are defined in terms of PGD.
- For bridges, fragility curves are defined in terms of Sa (0.3 sec), Sa(1.0 sec) and
PGD.
- For tunnels, fragility curves are defined in terms of PGA and PGD.

Definitions of various damage states and the methodology used in deriving all these
fragility curves are presented in the following sections.

7.1.5 Description of Highway Components

As mentioned previously, a highway system is composed of three components: roadways,
bridges and tunnels. In this section, a brief description of each is given.

Roadways

Roadways are classified as major roads and urban roads. Major roads include
interstate and state highways and other roads with four lanes or more. Parkways are
also classified as major roads. Urban roads include intercity roads and other roads
with two lanes.

Bridges
Bridges are classified based on the following structural characteristics:
- Seismic Design
- Number of spans: single vs. multiple span bridges
- Structure type: concrete, steel, others
- Pier type: multiple column bents, single column bents and pier walls
- Abutment type and bearing type: monolithic vs. non-monolithic; high rocker bearings,
low steel bearings and neoprene rubber bearings
- Span continuity: continuous, discontinuous (in-span hinges), and simply supported.

The seismic design of a bridge is taken into account in terms of the (i) spectrum
modification factor, (ii) strength reduction factor due to cyclic motion, (iii) drift limits,
and (iv) the longitudinal reinforcement ratio.
75
HazusMHTechnicalManual
This classification scheme incorporates various parameters that affect damage into
fragility analysis and provides a means to obtain better fragility curves when data become
available. A total of 28 classes (HWB1 through HWB28) are defined this way. These
classes differentiate between the different bridge characteristics found in the National
Bridge Inventory (NBI).
Tables 7.1.a and 7.1.b summarize the key NBI characteristics used, while Table 7.2
presents the 28 classes derived for Hazus. Please refer to Table 3.6 in Chapter 3 for the
full definitions of these bridges.
Table 7.1.a Bridge material Classes in NBI [NBI, 1988]
Code Description
1 Concrete
2 Concrete continuous
3 Steel
4 Steel continuous
5 Prestressed concrete
6 Prestressed concrete continuous
7 Timber
8 Masonry
9 Aluminium, Wrought Iron, or Cast Iron
0 Other
Table 7.1.b Bridge Types in NBI [NBI, 1988]
Code Description
01 Slab
02 Stringer/Multi-beam or Girder
03 Girder and Floor beam System
04 Tee Beam
05 Box Beam or Girders - Multiple
06 Box Beam or Girders single or Spread
07 Frame
08 Orthotropic
09 Truss Deck
10 Truss Thru
11 Arch Deck
12 Arch Thru
13 Suspension
14 Stayed Girder
15 Movable Lift
16 Movable Bascule
17 Movable Swing
18 Tunnel
19 Culvert
20 Mixed Types (applicable only to approach spans)
21 Segmental Box Girder
22 Channel Beam
00 Other

Table 7.2 Hazus Bridge Classification Scheme
76
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
CLASS NBI Class State Year Built
#
Spans
Length of
Max. Span
(meter)
Length
less than
20 m
K
3D

(See note
below)
I
shape

(See note
below)
Design Description
HWB1 All Non-CA <1990 >150 N/A EQ1 0 Conventional Major Bridge - Length >150m
HWB1 All CA <1975 >150 N/A EQ1 0 Conventional Major Bridge - Length >150m
HWB2 All Non-CA >=1990 >150 N/A EQ1 0 Seismic Major Bridge - Length >150m
HWB2 All CA >=1975 >150 N/A EQ1 0 Seismic Major Bridge - Length >150m
HWB3 All Non-CA <1990 1 N/A EQ1 1 Conventional Single Span
HWB3 All CA <1975 1 N/A EQ1 1 Conventional Single Span
HWB4 All Non-CA >=1990 1 N/A EQ1 1 Seismic Single Span
HWB4 All CA >=1975 1 N/A EQ1 1 Seismic Single Span
HWB5 101-106 Non-CA <1990 N/A EQ1 0 Conventional
Multi-Col. Bent, Simple Support -
Concrete
HWB6 101-106 CA <1975 N/A EQ1 0 Conventional
Multi-Col. Bent, Simple Support
- Concrete
HWB7 101-106 Non-CA >=1990 N/A EQ1 0 Seismic
Multi-Col. Bent, Simple Support
- Concrete
HWB7 101-106 CA >=1975 N/A EQ1 0 Seismic
Multi-Col. Bent, Simple Support
- Concrete
HWB8 205-206 CA <1975 N/A EQ2 0 Conventional
Single Col., Box Girder -
Continuous Concrete
HWB9 205-206 CA >=1975 N/A EQ3 0 Seismic
Single Col., Box Girder -
Continuous Concrete
HWB10 201-206 Non-CA <1990 N/A EQ2 1 Conventional Continuous Concrete
HWB10 201-206 CA <1975 N/A EQ2 1 Conventional Continuous Concrete
HWB11 201-206 Non-CA >=1990 N/A EQ3 1 Seismic Continuous Concrete
HWB11 201-206 CA >=1975 N/A EQ3 1 Seismic Continuous Concrete
HWB12 301-306 Non-CA <1990 No EQ4 0 Conventional
Multi-Col. Bent, Simple Support -
Steel
HWB13 301-306 CA <1975 No EQ4 0 Conventional
Multi-Col. Bent, Simple Support
- Steel
HWB14 301-306 Non-CA >=1990 N/A EQ1 0 Seismic
Multi-Col. Bent, Simple Support
- Steel
HWB14 301-306 CA >=1975 N/A EQ1 0 Seismic
Multi-Col. Bent, Simple Support
- Steel
HWB15 402-410 Non-CA <1990 No EQ5 1 Conventional Continuous Steel
HWB15 402-410 CA <1975 No EQ5 1 Conventional Continuous Steel
HWB16 402-410 Non-CA >=1990 N/A EQ3 1 Seismic Continuous Steel
HWB16 402-410 CA >=1975 N/A EQ3 1 Seismic Continuous Steel
77
HazusMHTechnicalManual
Table 7.2 Hazus Bridge Classification Scheme (Continued)
CLASS
NBI
Class
State
Year
Built
# Spans
Length of
Max. Span
(meter)
Length
less than
20 m
K
3D

(notes
below)
I
shape

(notes
below)
Design Description
HWB17 501-506 Non-CA <1990 N/A EQ1 0 Conventional
Multi-Col. Bent, Simple
Support - Prestressed
Concrete
HWB18 501-506 CA <1975 N/A EQ1 0 Conventional
Multi-Col. Bent, Simple
Support - Prestressed
Concrete
HWB19 501-506 Non-CA >=1990 N/A EQ1 0 Seismic
Multi-Col. Bent, Simple
Support - Prestressed
Concrete
HWB19 501-506 CA >=1975 N/A EQ1 0 Seismic
Multi-Col. Bent, Simple
Support - Prestressed
Concrete
HWB20 605-606 CA <1975 N/A EQ2 0 Conventional
Single Col., Box Girder -
Prestressed Continuous
Concrete
HWB21 605-606 CA >=1975 N/A EQ3 0 Seismic
Single Col., Box Girder -
Prestressed Continuous
Concrete
HWB22 601-607 Non-CA <1990 N/A EQ2 1 Conventional Continuous Concrete
HWB22 601-607 CA <1975 N/A EQ2 1 Conventional Continuous Concrete
HWB23 601-607 Non-CA >=1990 N/A EQ3 1 Seismic Continuous Concrete
HWB23 601-607 CA >=1975 N/A EQ3 1 Seismic Continuous Concrete
HWB24 301-306 Non-CA <1990 Yes EQ6 0 Conventional
Multi-Col. Bent, Simple
Support - Steel
HWB25 301-306 CA <1975 Yes EQ6 0 Conventional
Multi-Col. Bent, Simple
Support - Steel
HWB26 402-410 Non-CA <1990 Yes EQ7 1 Conventional Continuous Steel
HWB27 402-410 CA <1975 Yes EQ7 1 Conventional Continuous Steel
HWB28
All other bridges that are not
classified

EQ1 through EQ7 in Table 7.2 are equations for evaluating K
3D.
K
3D
is a factor that
modifies the piers 2-dimensional capacity to allow for the 3-dimensional arch action in
the deck. All of the equations have the same functional form; K
3D
=1 +A / (N B),
where N is the number of spans and the parameters A and B are given in table 7.3.

The I
shape
term (given in table 7.2) is a Boolean indicator. The K
shape
factor is the
modifier that converts cases for short periods to an equivalent spectral amplitude at T=1.0
second. When I
shape
=0, the K
shape
factor does not apply. When I
shape
=1, the K
shape
factor
applies. Later in this section, the use of the K
shape
factor will be illustrated through an
example.

The 28 bridge classes in Table 7.2 (HWB1 through HWB28) reflect the maximum
number of combinations for standard bridge classes. Attributes such as the skeweness
and number of spans are accounted for in the evaluation of damage potential through a
modification scheme that is presented later in this section.
78
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems

Table 7.3 Coefficients for Evaluating K
3D

Equation A B K
3D

EQ1 0.25 1 1 +0.25 / (N 1)
EQ2 0.33 0 1 +0.33 / (N)
EQ3 0.33 1 1 +0.33 / (N 1)
EQ4 0.09 1 1 +0.09 / (N 1)
EQ5 0.05 0 1 +0.05 / (N)
EQ6 0.20 1 1 +0.20 / (N 1)
EQ7 0.10 0 1 +0.10 / (N)


Tunnels

Tunnels are classified as bored/drilled or cut & cover.

7.1.6 Definitions of Damage States

A total of five damage states are defined for highway system components. These are
none (ds
1
), slight/minor (ds
2
), moderate (ds
3
), extensive (ds
4
) and complete (ds
5
).

Slight/Minor Damage (ds
2
)

- For roadways, ds
2
is defined by slight settlement (few inches) or offset of the
ground.

- For bridges, ds
2
is defined by minor cracking and spalling to the abutment,
cracks in shear keys at abutments, minor spalling and cracks at hinges, minor
spalling at the column (damage requires no more than cosmetic repair) or
minor cracking to the deck

- For tunnels, ds
2
is defined by minor cracking of the tunnel liner (damage
requires no more than cosmetic repair) and some rock falling, or by slight
settlement of the ground at a tunnel portal.


Moderate Damage (ds
3
)
- For roadways, ds
3
is defined by moderate settlement (several inches) or offset
of the ground.

79
HazusMHTechnicalManual
- For bridges, ds
3
is defined by any column experiencing moderate (shear cracks)
cracking and spalling (column structurally still sound), moderate movement of the
abutment (<2"), extensive cracking and spalling of shear keys, any connection
having cracked shear keys or bent bolts, keeper bar failure without unseating,
rocker bearing failure or moderate settlement of the approach.

- For tunnels, ds
3
is defined by moderate cracking of the tunnel liner and rock
falling.

Extensive Damage (ds
4
)
- For roadways, ds
4
is defined by major settlement of the ground (few feet).

- For bridges, ds
4
is defined by any column degrading without collapse shear
failure - (column structurally unsafe), significant residual movement at
connections, or major settlement approach, vertical offset of the abutment,
differential settlement at connections, shear key failure at abutments.

- For tunnels, ds
4
is characterized by major ground settlement at a tunnel portal
and extensive cracking of the tunnel liner.

Complete Damage (ds
5
)
- For roadways, ds
5
is defined by major settlement of the ground (i.e., same as
ds
4
).

- For bridges, ds
5
is defined by any column collapsing and connection losing all
bearing support, which may lead to imminent deck collapse, tilting of substructure
due to foundation failure.

- For tunnels, ds
5
is characterized by major cracking of the tunnel liner, which
may include possible collapse.

7.1.7 Component Restoration Curves

Restoration curves are developed based on a best fit to ATC-13 data for the social
function classifications of interest (SF 25a through SF 25e) consistent with damage states
defined in the previous section (first four classes in ATC-13). Figure 7.1 shows
restoration curves for urban and major roads, Figure 7.2 represents restoration curves for
highway bridges, while Figure 7.3 shows restoration curves for highway tunnels. The
smooth curves shown in these figures are normal curves characterized by a mean and a
standard deviation. The parameters of these restoration curves are given in Tables 7.4
and 7.5. The former table gives means and standard deviations for each restoration curve
(i.e., smooth continuous curve), while the second table gives approximate discrete
functions for the restoration curves developed.
710
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems

Table 7.4 Continuous Restoration Functions for Highways (after ATC-13, 1985)
Roadways Highway Bridges Highway Tunnels
Damage State Mean
(Days)
o
(days)
Mean
(Days)
o
(days)
Mean
(Days)
o
(days)
Slight/Minor 0.9 0.05 0.6 0.6 0.5 0.3
Moderate 2.2 1.8 2.5 2.7 2.4 2.0
Extensive
21 16
75.0 42.0 45.0 30.0
Complete

230.0 110.0 210.0 110.0

The values shown in Table 7.5 below represent distributions on functionality for each
restoration period based on damage state immediately after the earthquake.
Table 7.5 Discrete Restoration Functions for Highways
Roadways
Restoration Functional Percentage
Period Slight Moderate Extensive/Complete
1 day 90 25 10
3 days 100 65 14
7 days 100 100 20
30 days 100 100 70
90 days 100 100 100
Bridges
Restoration Functional Percentage
Period Slight Moderate Extensive Complete
1 day 70 30 2 0
3 days 100 60 5 2
7 days 100 95 6 2
30 days 100 100 15 4
90 days 100 100 65 10
Tunnels
Restoration Functional Percentage
Period Slight Moderate Extensive Complete
1 day 90 25 5 0
3 days 100 65 8 3
7 days 100 100 10 3
30 days 100 100 30 5
90 days 100 100 95 15

711
HazusMHTechnicalManual
7.1.8 Development of Damage Functions

Fragility curves for highway system components are defined with respect to classification
and ground motion parameter.

Damage functions for Roadways

Fragility curves for major roads and urban roads are shown in Figures 7.4. and
7.5, respectively. The medians and dispersions of these curves are presented in
Table 7.6.
Table 7.6 Damage Algorithms for Roadways
Permanent Ground Deformation
Components Damage State Median (in) |
slight/minor 12 0.7
Major Road moderate 24 0.7
(Hrd1) extensive/complete 60 0.7
slight/minor 6 0.7
Urban Roads moderate 12 0.7
(Hrd2) extensive/complete 24 0.7

Damage Functions for Bridges

There are 28 primary bridge types for which all four damage states are identified
and described. For other bridges, fragility curves of the 28 primary bridge types
are adjusted to reflect the expected performance of a specific bridge which may
be better or worse than the corresponding primary bridge type.

A total of 224 bridge damage functions are obtained, 112 due to ground shaking
and 112 due to ground failure. For a complete description on the theoretical
background of the damage functions, see Basoz and Mander (1999). This
document is referenced at the end of this section and can be obtained from NIBS
or FEMA.

Medians of these damage functions are given in Table 7.7. The dispersion is set
to 0.6 for the ground shaking damage algorithm and 0.2 for the ground failure
damage algorithm. Only incipient unseating and collapse (i.e., which correspond
to extensive and complete damage states) are considered as possible types of
damage due to ground failure. Initial damage to bearings (i.e., which would
correspond to slight and/or moderate damage states) from ground failure is not
considered.

Figures 7.6 and 7.7 show example fragility curves for major bridges.
712
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
Table 7.7 Damage Algorithms for Bridges

Sa [1.0 sec in gs] for Damage Functions
due to Ground Shaking
PGD [inches] for Damage Functions
due to Ground Failure
CLASS Slight Moderate Extensi ve Complete Slight Moderate Extensi ve Complete
HWB1 0.40 0.50 0.70 0.90 3.9 3.9 3.9 13.8
HWB2 0.60 0.90 1.10 1.70 3.9 3.9 3.9 13.8
HWB3 0.80 1.00 1.20 1.70 3.9 3.9 3.9 13.8
HWB4 0.80 1.00 1.20 1.70 3.9 3.9 3.9 13.8
HWB5 0.25 0.35 0.45 0.70 3.9 3.9 3.9 13.8
HWB6 0.30 0.50 0.60 0.90 3.9 3.9 3.9 13.8
HWB7 0.50 0.80 1.10 1.70 3.9 3.9 3.9 13.8
HWB8 0.35 0.45 0.55 0.80 3.9 3.9 3.9 13.8
HWB9 0.60 0.90 1.30 1.60 3.9 3.9 3.9 13.8
HWB10 0.60 0.90 1.10 1.50 3.9 3.9 3.9 13.8
HWB11 0.90 0.90 1.10 1.50 3.9 3.9 3.9 13.8
HWB12 0.25 0.35 0.45 0.70 3.9 3.9 3.9 13.8
HWB13 0.30 0.50 0.60 0.90 3.9 3.9 3.9 13.8
HWB14 0.50 0.80 1.10 1.70 3.9 3.9 3.9 13.8
HWB15 0.75 0.75 0.75 1.10 3.9 3.9 3.9 13.8
HWB16 0.90 0.90 1.10 1.50 3.9 3.9 3.9 13.8
HWB17 0.25 0.35 0.45 0.70 3.9 3.9 3.9 13.8
HWB18 0.30 0.50 0.60 0.90 3.9 3.9 3.9 13.8
HWB19 0.50 0.80 1.10 1.70 3.9 3.9 3.9 13.8
HWB20 0.35 0.45 0.55 0.80 3.9 3.9 3.9 13.8
HWB21 0.60 0.90 1.30 1.60 3.9 3.9 3.9 13.8
HWB22 0.60 0.90 1.10 1.50 3.9 3.9 3.9 13.8
HWB23 0.90 0.90 1.10 1.50 3.9 3.9 3.9 13.8
HWB24 0.25 0.35 0.45 0.70 3.9 3.9 3.9 13.8
HWB25 0.30 0.50 0.60 0.90 3.9 3.9 3.9 13.8
HWB26 0.75 0.75 0.75 1.10 3.9 3.9 3.9 13.8
HWB27 0.75 0.75 0.75 1.10 3.9 3.9 3.9 13.8
HWB28 0.80 1.00 1.20 1.70 3.9 3.9 3.9 13.8
713
HazusMHTechnicalManual

The damage algorithm for bridges can be broken into eight steps:

Step 1:
Get the bridge location (longitude and latitude), class (HWB1 through HWB28), number
of spans (N), skew angle (o), span width (W), bridge length (L), and maximum span
length (L
max
). Note that the skew angle is defined as the angle between the centerline of a
pier and a line normal to the roadway centerline.

Step 2:
Evaluate the soil-amplified shaking at the bridge site. That is, get the peak ground
acceleration (PGA), spectral accelerations (Sa[0.3 sec] and Sa[1.0 sec] ) and the
permanent ground deformation (PGD).

Step 3:
Evaluate the following three modification factors:

K
skew
=sqrt[sin(90-o)]

K
shape
=2.5 x Sa(1.0 sec) / Sa(0.3 sec)

K
3D
=1 +A / (N B) A and B are read from Table 7.3

Step 4:
Modify the ground shaking medians for the standard fragility curves in Table 7.7 as
follows:

New Median [for slight] =Old Median [for slight] x Factor
slight

Where
Factor
slight
=1 if I
shape
=0 (I
shape
is read from Table 7.2)
or
Factor
slight
=minimum of (1, K
shape
) if I
shape
=1

New median [moderate] =Old median [for moderate] * ( K
skew
) * ( K
3D
)


New median [extensive] =Old median [for extensive] * ( K
skew
) * ( K
3D
)


New median [complete] =Old median [for complete] * ( K
skew
) * ( K
3D
)


Step 5:
Use the new medians along with the dispersion | =0.6 to evaluate the ground shaking-
related damage state probabilities. Note that Sa(1.0 sec) (listed in Table 7.7) is the
parameter to use in this evaluation.

714
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
Step 6:
Modify the PGD medians for the standard fragility curves listed in Table 7.7 as
follows:

New PGD median [for slight] =Table7.7 PGD median [for slight] x f
1
New PGD median [moderate] =Table7.7 PGD median [for moderate] x f
1

New PGD median [extensive] =Table7.7 PGD median [for extensive] x f
1

New PGD median [complete] =Table7.7 median [for complete] x f
2

Where f
1
and f
2
are modification factors that are functions of the number of spans (N),
width of the span (W), length of the bridge (L), and the skewness (o) and can be
computed using the equations in Table 7.8 below.
Table 7.8 Modifiers for PGD Medians
CLASS f
1
f
2
HWB1 1 1
HWB2 1 1
HWB3 1 1
HWB4 1 1
HWB5 0.5 * L / [ N . W . sin (o) ] 0.5 * L / [ N . W . sin (o) ]
HWB6 0.5 * L / [ N . W . sin (o) ] 0.5 * L / [ N . W . sin (o) ]
HWB7 0.5 * L / [ N . W . sin (o) ] 0.5 * L / [ N . W . sin (o) ]
HWB8 1 sin (o)
HWB9 1 sin (o)
HWB10 1 sin (o)
HWB11 1 sin (o)
HWB12 0.5 * L / [ N . W . sin (o) ] 0.5 * L / [ N . W . sin (o) ]
HWB13 0.5 * L / [ N . W . sin (o) ] 0.5 * L / [ N . W . sin (o) ]
HWB14 0.5 * L / [ N . W . sin (o) ] 0.5 * L / [ N . W . sin (o) ]
HWB15 1 sin (o)
HWB16 1 sin (o)
HWB17 0.5 * L / [ N . W . sin (o) ] 0.5 * L / [ N . W . sin (o) ]
HWB18 0.5 * L / [ N . W . sin (o) ] 0.5 * L / [ N . W . sin (o) ]
HWB19 0.5 * L / [ N . W . sin (o) ] 0.5 * L / [ N . W . sin (o) ]
HWB20 1 sin (o)
HWB21 1 sin (o)
HWB22 0.5 * L / [ N . W . sin (o) ] 0.5 * L / [ N . W . sin (o) ]
HWB23 0.5 * L / [ N . W . sin (o) ] 0.5 * L / [ N . W . sin (o) ]
HWB24 0.5 * L / [ N . W . sin (o) ] 0.5 * L / [ N . W . sin (o) ]
HWB25 0.5 * L / [ N . W . sin (o) ] 0.5 * L / [ N . W . sin (o) ]
HWB26 1 sin (o)
HWB27 1 sin (o)
HWB28 1 1


715
HazusMHTechnicalManual

Step 7:
Use the new medians along with the dispersion | =0.2 to evaluate ground failure-related
damage state probabilities.

Step 8:
Combine the damage state probabilities and evaluate functionality of bridge.

Example of bridge damage evaluation:
Consider a three-span simply supported prestressed concrete bridge seated on neoprene
bearings located in the Memphis area. The table below lists the data for this bridge
obtained from NBI. For the scenario earthquake, assume that the ground motion for rock
conditions (NEHRP class B) is defined by the following parameters:

Sa(0.3 sec) =2.1g, Sa(1.0 sec) =0.24g PGA =0.38g

Also, assume that the bridge is located in soil type D.

The median spectral acceleration ordinates for different damage states are determined as
follows:

Step 1: Ground motion data is amplified for soil conditions (Table 4.10 in Chapter 4):

Sa(0.3 sec) =2.1g (1 x 2.1g),
Sa(1.0 sec) =0.43g (1.8 x 0.24g)
PGA =0.53g (1.4 x 0.38g )

Step 2: The bridge class is determined.
Bridge Data Required for the Analysis
NBI field Data Remarks
27 1968 Year built
34 32 Angle of skew
43 501 Prestressed concrete, simple span
45 3 Number of spans
48 23 Maximum span length (m)
49 56 Total bridge length (m)

Based on the information in the table above, HWB17 is determined to be the bridge
class.

Step 3: Parameters needed in evaluating the median spectral accelerations are
computed:

K
skew
=sqrt[sin(90-o)] =sqrt [sin (90 32) ] =0.91
716
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems

K
shape
=2.5 x Sa(1.0 sec) / Sa(0.3 sec) =0.50

K
3D
=1 +A / (N B) =1 +0.25 / (3-1) =1.125 (See Tables 7.2 and 7.3)

Step 4:

From Table 7.2, I
shape
is 0 for HWB17, therefore long periods governs, and
Factor
slight
is 1. Therefore:

New Sa[1.0 sec] [for slight] =Old Sa[1.0 sec] [for slight] *Factor
slight
=0.25g * 1.00 =0.25g
New Sa[1.0 sec] [moderate] =Old Sa[1.0 sec] [for moderate] * ( K
skew
) * ( K
3D
)
=0.35g * 0.91 * 1.125 =0.36g
New Sa[1.0 sec] [extensive] =Old Sa[1.0 sec] [for extensive] * ( K
skew
) * ( K
3D
)

=0.45g * 0.91 * 1.125 =0.46g
New Sa[1.0 sec] [complete] =Old Sa[1.0 sec] [for complete] * ( K
skew
) * ( K
3D
)

=0.70g * 0.91 * 1.125 =0.72g

Step 5:
With these new medians, the shaking-related discrete damage state probabilities are
(using lognormal functions with the above medians and with betas equal to 0.6):

P[No damage] =1 0.82 =0.18
P[Slight damage] =0.82 0.62 =0.20
P[Moderate damage] =0.62 0.46 =0.16
P[Extensive damage] =0.46 0.20 =0.26
P[Complete damage] =0.20

Damage Functions for Tunnels

The tunnel damage functions developed based on the damage potential of their
subcomponents, namely the liner and the portal (G&E, 1994). G&E findings are based
partly on earthquake experience data reported by Dowding et. al. (1978) and Owen et. al
(1981). The subcomponent damage functions are given in Tables A.7.1 and A.7.2.

Ten tunnel damage functions were developed, four damage functions for ground shaking
(PGA) and six damage functions for ground failure (PGD). Medians and dispersion
factors of these damage functions are given in Table 7.9. Graphical representations of
these damage functions are also provided. Figures 7.8 and Figure 7.9 plot fragility curves
due to PGA for bored/drilled and cut & cover tunnels, respectively, while Figure 7.10
presents fragility curves for tunnels due to PGD.

717
HazusMHTechnicalManual
Table 7.9 Damage Algorithms for Tunnels
Peak Ground Acceleration
Classification Damage State Median (g) |
Bored/Drilled
(HTU1)
slight/minor
moderate
0.6
0.8
0.6
0.6
Cut & Cover
(HTU2)
slight/minor
moderate
0.5
0.7
0.6
0.6
Permanent Ground Deformation
Classification Damage State Median (in) |
Bored/Drilled
(HTU1)
slight/moderate
extensive
complete
6.0
12.0
60.0
0.7
0.5
0.5
Cut & Cover
(HTU2)
slight/moderate
extensive
complete
6.0
12.0
60.0
0.7
0.5
0.5

7.1.9 Guidance for Loss Estimation Using Advanced Data and Models Analysis

For an advanced analysis, experts can use the methodology developed with the flexibility
to (1) include a more refined inventory of the transportation system pertaining to the
study area, and (2) include component-specific and system-specific fragility data. User-
supplied damage algorithms can be modified to incorporate improved information about
key components of the highway system. Similarly, improved restoration curves can be
developed, given knowledge of available resources and a more accurate layout of the
transportation network within the local topographic (i.e., if the redundancy and
importance of highway components of the network are known).

7.1.10 References
(1) Applied Technology Council, "Earthquake Damage Evaluation Data for California", ATC-13,
Redwood City, CA, 1985.
(2) Dowding, C.H. and Rozen, A., "Damage to Rock Tunnels from Earthquake Shaking", Journal
of the Geotechnical Engineering Division, American Society of Civil Engineers, New York, NY,
February 1978.
(3) National Institute of Building Sciences, Enhancement of the Highway Transportation
Lifeline Module in Hazus, prepared by Nesrin Basoz and J ohn Mander, J anuary 1999.
(4) Kim, S.H., "A GIS-Based Regional Risk Analysis Approach for Bridges against Natural
Hazards", a dissertation submitted to the faculty of the graduate school of the State University of
New York at Buffalo, September 1993.
(5) Owen, G.N. and Scholl, R.E., "Earthquake Engineering Analysis of a Large Underground
Structures", Federal Highway Administration and National Science Foundation, FHWA/RD-
80/195, J anuary 1981.
(6) G&E Engineering Systems, Inc. (G&E), "NIBS Earthquake Loss Estimation Methods,
Technical Manual, Transportation Systems (Highway Systems)", May 1994.

718
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems



Time (days)

P
e
r
c
e
n
t

F
u
n
c
t
i
o
n
a
l

0
25
50
75
100
1 10 100 1000
Minor Moderate Extensive/Complete ATC-13 Data

Figure 7.1 Restoration Curves for Urban and Major Roads (after ATC-13, 1985).

Time (days)

P
e
r
c
e
n
t

F
u
n
c
t
i
o
n
a
l

0
25
50
75
100
1 10 100 1000
Minor Moderate Extensive Complete ATC-13 Data


Figure 7.2 Restoration Curves for Highway Bridges (after ATC-13, 1985).



719
HazusMHTechnicalManual








Time (days)

P
e
r
c
e
n
t

F
u
n
c
t
i
o
n
a
l

0
25
50
75
100
1 10 100 1000
Minor Moderate Extensive Complete ATC-13 Data


Figure 7.3 Restoration Curves for Highway Tunnels (after ATC-13, 1985).













720
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
Permanent Ground Deformation (inches)
[

P
r
o
b
a
b
i
l
i
t
y


D
s


>


d
s


|


P
G
D

]
0.00
0.25
0.50
0.75
1.00
0.00 6.00 12.00 18.00 24.00 30.00 36.00 42.00 48.00
Minor Moderate Extensive/Complete


Figure 7.4 Fragility Curves at Various Damage States for Interstate and State
Highways.


Permanent Ground Deformation (inches)
[

P
r
o
b
a
b
i
l
i
t
y


D
s


>


d
s


|


P
G
D

]
0.00
0.25
0.50
0.75
1.00
0.00 6.00 12.00 18.00 24.00 30.00 36.00 42.00 48.00
Minor Moderate Extensive/Complete

Figure 7.5 Fragility Curves at Various Damage States for Urban roads.
721
HazusMHTechnicalManual



Figure 7.6 Fragility Curves for Conventially Designed Major Bridges (HWB1).



Figure 7.7 Fragility Curves for Seismically Designed Major Bridges (HWB2).




0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6
D
a
m
a
g
e

S
t
a
t
e

E
x
c
e
e
d
i
n
g

P
r
o
b
a
b
i
l
i
t
y

Spectral Acceleration at 1.0 sec [g's]
Prob >=slight
Prob >=Mod
Prob >=Ext
Prob >=Comp
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6
D
a
m
a
g
e

S
t
a
t
e

E
x
c
e
e
d
i
n
g

P
r
o
b
a
b
i
l
i
t
y

Spectral Acceleration at 1.0 sec [g's]
Prob >=slight
Prob >=Mod
Prob >=Ext
Prob >=Comp
722
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
Peak Ground Acceleration (g)
P
r
o
b
a
b
i
l
i
t
y

[

D
s


>


d
s


|


P
G
A

]








0.0000
0.2500
0.5000
0.7500
1.0000
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Slight/Minor Moderate

Figure 7.8 Fragility Curves at Various Damage States for Bored/Drilled Tunnels
Subject to Peak Ground Acceleration.

Peak Ground Acceleration (g)
P
r
o
b
a
b
i
l
i
t
y

[

D
s


>


d
s


|


P
G
A

]








0.0000
0.2500
0.5000
0.7500
1.0000
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Slight/Minor Moderate

Figure 7.9 Fragility Curves at Various Damage States for Cut & Cover Tunnels
Subject to Peak Ground Acceleration.







723
HazusMHTechnicalManual







Permanent Ground Deformation (inches)
P
r
o
b
a
b
i
l
i
t
y

[

D
s


>


d
s


|


P
G
D

]









0
0.25
0.5
0.75
1
0.00 6.00 12.00 18.00 24.00 30.00 36.00 42.00 48.00
Slight/Minor Moderate Extensive / Complete




Figure 7.10 Fragility Curves at Various Damage States for All Types of Tunnels
Subject to Permanent Ground deformation.
724
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
7.2 Railway Transportation System

7.2.1 Introduction

This section presents an earthquake loss estimation methodology for a railway
transportation system. This system consists of tracks/roadbeds, bridges, tunnels, urban
stations, maintenance facilities, fuel facilities, and dispatch facilities. Past earthquake
damage reveals that bridges, tunnels, urban stations, maintenance facilities, fuel facilities,
and dispatch facilities are vulnerable to both ground shaking and ground failure, while
railway tracks/roadbeds are significantly affected by ground failure alone. Railway
tracks located on soft soil or fill or which cross a surface fault rupture can experience
failure resulting in loss of functionality. Railway bridges that fail usually result in
significant disruption to the transportation network, especially bridges that cross
waterways. Likewise, railway tunnels are often not redundant, and major disruption to
the transportation system is likely to occur should a tunnel become non-functional.

7.2.2 Scope

The scope of this section includes development of methods for estimation of earthquake
damage to a railway transportation system given knowledge of the system's components
(i.e., tracks, bridges, tunnels, stations, maintenance facilities, fuel facilities, or dispatch
facilities), the classification of each component (e.g., for fuel facilities, whether the
equipment within the facility is anchored or not), and the ground motion (i.e. peak ground
acceleration and permanent ground deformation).

Damage states describing the level of damage to each railway system component are
defined (i.e. slight/minor, moderate, extensive or complete). Damage states are related to
damage ratio (defined as ratio of repair to replacement cost) for evaluation of direct
economic loss. Component restoration curves are provided for each damage state to
evaluate loss of function. Restoration curves describe the fraction or percentage of the
component that is expected to be open or operational as a function of time following the
earthquake. For example, an extensively damaged railway facility might be closed (0%
functional) immediately following the earthquake, but 100% functional after 30 days.

Fragility curves are developed for each type of railway system component. These curves
describe the probability of reaching or exceeding each damage state given the level of
ground motion.

Evaluation of component functionality is done similar to the way it was done for highway
components.

Interdependence of components on the overall system functionality is not addressed by
the methodology. Such considerations require a system (network) analysis.



725
HazusMHTechnicalManual
7.2.3 Input Requirements and Output Information

Required input to estimate damage to railway systems includes the following items:

Track and Roadbeds

- Geographical location of railway links [longitude and latitude of end nodes]
- Permanent ground deformation (PGD) at trackbed link

Railway Bridges

- Geographical location of bridge (longitude and latitude)
- Spectral Acceleration at 0.3 and 1.0 seconds and PGD at bridge
- Bridge classification

Railway Tunnels

- Geographical location of tunnels (longitude and latitude)
- PGA and PGD at tunnel
- Tunnel classification

Railway System Facilities

- Geographical location of facilities (longitude and latitude)
- PGA and PGD at facility
- Facility classification

Direct damage output for railway systems includes probability estimates of (1)
component functionality and (2) physical damage, expressed in terms of the component's
damage ratio. Damage ratios, used as inputs to the direct economic loss module, are
presented in section 15.3 of Chapter 15.

Component functionality is described similar to highway system components, that is, by
the probability of being in a damage state (immediately following the earthquake) and by
the associated fraction or percentage of the component that is expected to be functional
after a specified period of time.

7.2.4 Form of Damage Functions

Damage functions or fragility curves for all railway system components described below
are modeled as lognormal functions that give the probability of reaching or exceeding
different levels of damage for a given level of ground motion. Each fragility curve is
characterized by a median value of ground motion (or failure) and an associated
dispersion factor (lognormal standard deviation). Ground motion is quantified in terms
of peak ground acceleration (PGA) and ground failure is quantified in terms of permanent
ground displacement (PGD).
726
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems

- For tracks/roadbeds, fragility curves are defined in terms of PGD.
- For railway bridges, fragility curves are defined similar to those for highway bridges
- For tunnels, fragility curves are the same as defined for highway systems (in terms of
PGA and PGD)
- For railway system facilities, fragility curves are defined in terms of PGA or Sd and
PGD.

Definitions of various damage states and the methodology used in deriving all these
fragility curves are presented in the following sections.

7.2.5 Description of Railway System Components

A railway system consists of four components: tracks/roadbeds, bridges, tunnels, and
facilities. This section provides a brief description of each.

Tracks/Roadbeds

Tracks/roadbeds refers to the assembly of rails, ties, and fastenings, and the ground
on which they rest. Only one classification is adopted for these components. This
classification is analogous to that of urban roads in highway systems.

Bridges

Railway bridges are classified similar to highway steel and concrete bridges.

Tunnels

Railway tunnels follow the same classification as highway tunnels. That is, they are
classified either as bored/drilled tunnels, or cut & cover tunnels.

Railway System Facilities

Railway system facilities include urban and suburban stations, maintenance facilities,
fuel facilities, and dispatch facilities.

Urban and Suburban stations: are generally key connecting hubs that are
important for system functionality. In western US, these buildings are mostly
made of reinforced concrete shear walls or moment resisting steel frames, while
in the eastern US, the small stations are mostly wood and the large ones are
mostly masonry or braced steel frames..

Maintenance facilities are housed in large structures that are not usually critical
for system functionality as maintenance activities can be delayed or performed
elsewhere. These building structures are often made of steel braced frames.

727
HazusMHTechnicalManual
Fuel facilities include buildings, tanks (anchored, unanchored, or buried), backup
power systems (if available, anchored or unanchored diesel generators), pumps,
and other equipment (anchored or unanchored). It should be mentioned that
anchored equipment in general refers to equipment designed with special seismic
tiedowns or tiebacks, while unanchored equipment refers to equipment designed
with no special considerations other than the manufacturer's normal requirements.
While some vibrating components, such as pumps, are bolted down regardless of
concern for earthquakes, as used here anchored means all components have
been engineered to meet seismic criteria which may include bracing (e.g., pipe or
stack bracing) or flexibility requirements (e.g., flexible connections across
separation joints) as well as anchorage. These definitions of anchored and
unanchored apply to all lifeline components. The fuel facility functionality is
determined with a fault tree analysis considering redundancies and subcomponent
behavior. Note that generic building damage functions are used in this fault tree
analysis for developing the overall fragility curve of fuel facilities. Above ground
tanks are typically made of steel with roofs also made of steel. Buried tanks are
typically concrete wall construction with concrete roofs. In total, five types of
fuel facilities are considered. These are: fuel facilities with or without anchored
equipment and with or without backup power (all combinations), and fuel
facilities with buried tanks.

Dispatch facilities consist of buildings, backup power supplies (if available,
anchored or unanchored diesel generators), and electrical equipment (anchored or
unanchored). Generic reinforced concrete building with shear walls damage
functions, are used in this fault tree analysis for developing the overall fragility
curves for dispatch facilities. In total, four types of dispatch facilities are
considered. These are dispatch facilities with or without anchored equipment and
with or without backup power (all combinations).

7.2.6 Definitions of Damage States

A total of five damage states are defined for railway system components. These are none
(ds
1
), slight/minor (ds
2
), moderate (ds
3
), extensive (ds
4
) and complete (ds
5
).

Slight/Minor Damage (ds
2
)

- For tracks and roadbeds, ds
2
is defined by minor (localized) derailment due to
slight differential settlement of embankment or offset of the ground.

- For railway bridges, ds
2
is defined similar to highway bridges.

- For railway tunnels, ds
2
is defined similar to highway tunnels.

- For railway system facilities,

728
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
^ for urban stations and maintenance facilities, ds
2
is defined by slight building
damage (check building module for full description of potential damage).

^ for fuel facilities with anchored equipment, ds
2
is defined by slight damage to
pump building, minor damage to anchor of tanks, or loss of off-site power (check
electric power systems for more on this) for a very short period and minor damage
to backup power (i.e. to diesel generators, if available).

^ for fuel facilities with unanchored equipment, ds
2
is defined by elephant foot
buckling of tanks with no leakage or loss of contents, slight damage to pump
building, or loss of commercial power for a very short period and minor damage
to backup power (i.e to diesel generators, if available).

^ for fuel facilities with buried tanks (PGD related damage), ds
2
is defined by
minor uplift (few inches) of the buried tanks or minor cracking of concrete walls.

^ for dispatch facilities with anchored equipment, ds
2
is defined by minor anchor
damage, slight damage to building, or loss of commercial power for a very short
period and minor damage to backup power (i.e. diesel generators, if available).

^ for dispatch facilities with unanchored equipment, ds
2
is defined by loss of off-
site power for a very short period and minor damage to backup power (i.e. to
diesel generators, if available), or slight damage to building.

Moderate Damage (ds
3
)

- For railway tracks and roadbeds, ds
3
is defined by considerable derailment due to
differential settlement or offset of the ground. Rail repair is required.

- For railway bridges, ds
3
is defined as for highway bridges.

- For railway tunnels, ds
3
is defined as for highway tunnels

- For railway system facilities,

^ for urban stations and maintenance facilities, ds
3
is defined by moderate
building damage (check building module for description of potential damage).

^ for fuel facilities with anchored equipment, ds
3
is defined by elephant foot
buckling of tanks with no leakage or loss of contents, considerable damage to
equipment, moderate damage to pump building, or loss of commercial power for
few days and malfunction of backup power (i.e., diesel generators, if available).

729
HazusMHTechnicalManual
^ for fuel facilities with unanchored equipment, ds
3
is defined by elephant foot
buckling of tanks with partial loss of contents, moderate damage to pump
building, loss of commercial power for few days and malfunction of backup
power (i.e., diesel generators, if available).

^ for fuel facilities with buried tanks, ds
3
is defined by damage to roof supporting
columns, and considerable cracking of walls.

^ for dispatch facilities with anchored equipment, ds
3
is defined by considerable
anchor damage, moderate damage to building, or loss of commercial power for
few days and malfunction of backup power (i.e., diesel generators, if available).

^ for dispatch facilities with unanchored equipment, ds
3
is defined by moderate
damage to building, or loss of off-site power for few days and malfunction of
backup power (i.e., diesel generators, if available)..

Extensive Damage (ds
4
)

- For railway tracks/roadbeds, ds
4
is defined by major differential settlement of the
ground resulting in potential derailment over extended length.

- For railway bridges, ds
4
is defined as for highway bridges.

- For railway tunnels, ds
4
is defined as for highway tunnels.

- For railway system facilities,

^ for urban stations and maintenance facilities, ds
4
is defined by extensive
building damage (check building module for description of potential damage).

^ for fuel facilities with anchored equipment, ds
4
is defined by elephant foot
buckling of tanks with loss of contents, extensive damage to pumps
(cracked/sheared shafts), or extensive damage to pump building.

^ for fuel facilities with unanchored equipment, ds
4
is defined by weld failure at
base of tank with loss of contents, extensive damage to pump building, or
extensive damage to pumps (cracked/sheared shafts).

^ for fuel facilities with buried tanks, ds
4
is defined by considerable uplift (more
than a foot) of the tanks and rupture of the attached piping.

^ For dispatch facilities with unanchored or anchored equipment, ds
4
is defined
by extensive building damage.
730
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems

Complete Damage (ds
5
)

- For railway tracks/roadbeds, ds
5
is the same as ds
4
.

- For railway bridges, ds
5
is defined as for highway bridges.

- For railway tunnels, ds
5
is defined as for highway tunnels.

- For railway system facilities,

^ For urban stations and maintenance facilities, ds
5
is defined by extensive to
complete building damage (check building module for description of potential
damage).

^ For fuel facilities with anchored equipment, ds
5
is defined by weld failure at
base of tank with loss of contents, or extensive to complete damage to pump
building.

^ For fuel facilities with unanchored equipment, ds
5
is defined by tearing of tank
wall or implosion of tank (with total loss of content), or extensive/complete
damage to pump building.

^ For fuel facilities with buried tanks, ds
5
is same as ds
4
.

^ For dispatch facilities with unanchored or anchored equipment, ds
5
is defined
by complete damage to building.


7.2.7 Component Restoration Curves

Restoration curves are developed based in part on ATC-13 damage data for the social
function classifications of interest (SF 26a through SF 26d) consistent with damage states
defined in the previous section. Normally distributed functions are used to approximate
these restoration curves, as was done for highway systems. Means and dispersions
(standard deviations) of these restoration functions are given in Table 7.10.a. Table
7.10.b gives approximate discrete functions for these developed restoration functions.
Figures 7.11 through 7.14 show restoration functions for railway tracks/roadbed, bridges,
tunnels and facilities, respectively. ATC-13 restoration data for railway terminal stations
are used to generically represent all other railway facilities.

731
HazusMHTechnicalManual
Table 7.10.a Continuous Restoration Functions for Railway System Components
(after ATC-13, 1985)
Classification Damage State Mean (Days) o (days)

Railway Tracks
slight/minor
moderate
extensive
complete
0.9
3.3
15.0
65.0
0.07
3.0
13.0
45.0

Railway Bridges
slight/minor
moderate
extensive
complete
0.9
2.8
31.0
110.0
0.06
1.8
22.0
73.0

Railway Tunnels
slight/minor
moderate
extensive
complete
0.9
4.0
37.0
150.0
0.05
3.0
30.0
80.0

Railway Facilities
slight/minor
moderate
extensive
complete
0.9
1.5
15.0
65.0
0.05
1.5
15.0
50.0




Table 7.10.b Discretized Restoration Functions for Railway System Components

Classification Damage State
1 day 3 days 7 days 30 days 90 days


Functional Percentage

Railway Tracks
slight/minor
moderate
extensive
complete
90
22
14
6
100
46
18
8
100
90
28
10
100
100
87
22
100
100
100
70

Railway Bridges
slight/minor
moderate
extensive
complete
80
15
9
7
100
55
10
7
100
100
14
8
100
100
50
14
100
100
100
40

Railway Tunnels
slight/minor
moderate
extensive
complete
95
16
11
3
100
38
13
4
100
85
16
4
100
100
40
7
100
100
97
22

Railway Facilities
slight/minor
moderate
extensive
complete
95
37
15
10
100
85
20
11
100
100
29
12
100
100
83
25
100
100
100
70


732
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
7.2.8 Development of Damage Functions

Fragility curves for railway system components are defined with respect to classification
and ground motion parameter.

Damage functions for Railway Tracks/Roadbeds

Damage functions for tracks/roadbeds are similar to those of major roads. The
medians and dispersions of these curves were given in Table 7.6 (see highway system
section).

Damage Functions for Railway Bridges

Fragility curves for bridges are the same ones presented in section 7.1.8 for highway
bridges.

Damage Functions for Tunnels

Tunnel damage functions are the same as those derived for highways. These were
given in Table 7.9 and plotted in Figures 7.9 and 7.10 of the "highway systems"
section.


Damage Functions for Railway System Facilities

Damage functions for railway system facilities are defined in terms of spectral
values and PGD. Note that, unless it is specified otherwise, ground failure
(PGD) related damage functions for these facilities are assumed to be similar to
those described for buildings. These are:

- For lateral spreading, a lognormal damage function with a median of 60 inches
and a dispersion of 1.2 is assumed for the damage state of "at least extensive".
20% of this damage is assumed to be complete. That is, for a PGD of 10" due to
lateral spreading, there is a 7% probability of "at least extensive" damage.

- For vertical settlement, a lognormal curve with a median of 10 inches and a
dispersion of 1.2 is assumed for the damage state of "at least extensive". 20% of
this damage is assumed to be complete. That is, for a PGD of 10" due to vertical
settlement, there is a 50% chance of "at least extensive" damage.

- For fault movement or landslide, a lognormal curve with a median of 10 inches
and a dispersion of 0.5 is assumed for "complete" damage state. That is, for 10
inches of PGD due to fault movement or landslide, there is a 50% chance of
"complete" damage.

733
HazusMHTechnicalManual
An example of how to combine multiple PGD algorithms with a PGA algorithm is
presented later in this section.

Damage Functions for Urban Stations and Maintenance Facilities

These damage functions are similar to the building fragility curves developed in
Chapter 5.


Damage Functions for Fuel Facilities

Fragility curves are developed for the five types of fuel facilities mentioned before,
namely, fuel facilities with anchored equipment and backup power, fuel facilities with
anchored equipment but no backup power, fuel facilities with unanchored equipment
and backup power, fuel facilities with unanchored equipment and no backup power,
and fuel facilities with buried tanks. Medians and dispersions of damage functions to
fuel facility subcomponents are summarized in Tables B.7.3 and B.7.4 of Appendix
7B. A generic building type is used in developing fragility curves for fuel facilities in
the specified fault tree logic (see Table B.7.3 of Appendix 7B). Note that the
interaction effects, specifically that of the electric power module, are considered in
this fault tree logic for the slight/minor and moderate damage states (refer to section
8.5.8 of Chapter 8 for more details on loss of commercial power effects on other
lifelines).

Component fragility curves are obtained using the same methodology as used for
bridges wherein a lognormal curve that best fits the results of the Boolean
combination is determined numerically. It should be mentioned that the Boolean
logic is implicitly presented within the definition of a particular damage state.

The fault tree shown in Figure 7.19a presents the Boolean logic for the case of
moderate damage to fuel facilities with anchored equipment and backup power, while
Figure 7.19b compares the fragility curve resulting from the Boolean combination to
the fitted lognormal fragility curve. The dotted line in Figure 7.19 represents the
overall fuel facility fragility curve.

The medians and dispersions of the damage functions for anchored and unanchored
fuel facilities are shown in Table 7.11. These damage functions are also shown as
fragility curves in Figures 7.20.a through 7.20e.

734
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
Table 7.11 Damage Algorithms for Fuel Facilities

Peak Ground Acceleration
Classification Damage State Median (g) |
Facility with
Anchored
Components w/
Backup Power
slight/minor
moderate
extensive
complete
0.23
0.43
0.64
1.10
0.50
0.45
0.60
0.60
Facility with
Anchored
Components w/o
Backup Power
slight/minor
moderate
extensive
complete
0.12
0.27
0.64
1.10
0.55
0.50
0.60
0.60
Facility with
Unnchored
Components w/
Backup Power
slight/minor
moderate
extensive
complete
0.10
0.23
0.48
0.80
0.55
0.50
0.60
0.60
Facility with
Unnchored
Components w/o
Backup Power
slight/minor
moderate
extensive
complete
0.09
0.20
0.48
0.80
0.50
0.45
0.60
0.60

Permanent Ground Deformation
Classification Damage State Median (in) |
Fuel facility w/
buried tanks
slight/minor
moderate
extensive/
Complete
4
8
24
0.5
0.5
0.5

PGA Related Damage Functions for Dispatch Facilities

As with fuel facilities, the same generic building type is used in developing the PGA
related fragility curves for dispatch facilities in the fault tree logic. The medians and
dispersions of the PGA related damage functions for anchored and unanchored
dispatch facilities are given in Table 7.12 and plotted in Figures 7.21.c through
7.21.d. Note that the medians and dispersions of the damage functions for dispatch
facility subcomponents are summarized in Tables B.7.5 and B.7.6 of Appendix 7B.

735
HazusMHTechnicalManual
Table 7.12 Damage Algorithms for Dispatch Facilities

Peak Ground Acceleration
Classification Damage State Median (g) |
Facility with
Anchored
Components w/
Backup Power
slight/minor
moderate
extensive
complete
0.15
0.35
0.8
1.50
0.75
0.65
0.80
0.80
Facility with
Anchored
Components w/o
Backup Power
slight/minor
moderate
extensive
complete
0.12
0.27
0.80
1.50
0.50
0.45
0.80
0.80
Facility with
Unnchored
Components w/
Backup Power
slight/minor
moderate
extensive
complete
0.13
0.28
0.80
1.50
0.55
0.50
0.80
0.80
Facility with
Unnchored
Components w/o
Backup Power
slight/minor
moderate
extensive
complete
0.11
0.23
0.80
1.50
0.45
0.40
0.80
0.80

Note that the values of Table 7.12 indicate that the damage functions of dispatch facilities
are mostly dominated by the building behavior.

Multiple Hazards Analysis for Railway System Facilities

In this section, a hypothetical example illustrating the methodology for combining
multiple hazards for nodal facilities is presented.

Assume that due to some earthquake, a railway fuel facility with anchored components
and backup power is subject to a PGA level of 0.3g, a lateral spreading displacement of
12 inches, a vertical settlement of 3 inches, and a potential landslide displacement of 15
inches. Assume also that the probability of liquefaction is 0.6, and that the probability of
landslide is 0.7.

- Due to ground shaking, the following probabilities of exceedence are obtained:

P[ D
s
> ds
2
| PGA =0.3g ] =0.70
P[ D
s
> ds
3
| PGA =0.3g ] =0.21
P[ D
s
> ds
4
| PGA =0.3g ] =0.10
P[ D
s
> ds
5
| PGA =0.3g ] =0.02

- Due to vertical settlement, the following probabilities of exceedence are obtained:

P[ D
s
> ds
2
| PGD =3 inches ] =0.16
P[ D
s
> ds
3
| PGD =3 inches ] =0.16
736
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
P[ D
s
> ds
4
| PGD =3 inches] =0.16
P[ D
s
> ds
5
| PGD =3 inches ] =20% * 0.16 =0.03

- Due to lateral spreading, the following probabilities of exceedence are obtained:

P[ D
s
> ds
2
| PGD =12 inches ] =0.09
P[ D
s
> ds
3
| PGD =12 inches ] =0.09
P[ D
s
> ds
4
| PGD =12 inches] =0.09
P[ D
s
> ds
5
| PGD =12 inches ] =20% * 0.09 =0.02

Therefore, for liquefaction, vertical settlement controls

- Due to landslide, the following probabilities of exceedence are obtained:

P[ D
s
> ds
2
| PGD =15 inches ] =0.64
P[ D
s
> ds
3
| PGD =15 inches ] =0.64
P[ D
s
> ds
4
| PGD =15 inches] =0.64
P[ D
s
> ds
5
| PGD =15 inches ] =0.64

Next, we compute the combined probabilities of exceedence (from complete to
slight/minor):

P[ D
s
> ds
5
] =0.02 +0.6x0.03 +0.7x0.64
- 0.02x0.6x0.03 - 0.02x0.7x0.64 - 0.6x0.03x0.7x0.64
+0.02x0.6x0.03x0.7x0.64
=0.47
P[ D
s
> ds
4
] =0.10 +0.6x0.16 +0.7x0.64
- 0.10x0.6x0.16 - 0.10x0.7x0.64 - 0.6x0.16x0.7x0.64
+0.10x0.6x0.16x0.7x0.64
=0.55
P[ D
s
> ds
3
] =0.21 +0.6x0.16 +0.7x0.64
- 0.21x0.6x0.16 - 0.21x0.7x0.64 - 0.6x0.16x0.7x0.64
+0.21x0.6x0.16x0.7x0.64
=0.61
P[ D
s
> ds
2
] =0.70 +0.6x0.16 +0.7x0.64
- 0.70x0.6x0.16 - 0.16x0.7x0.64 - 0.6x0.16x0.7x0.64
+0.70x0.6x0.16x0.7x0.64
=0.85

Therefore, the combined discrete damage states probabilities are:

P[ D
s
= ds
1
] =1 - 0.85 =0.15
P[ D
s
= ds
2
] =0.85 - 0.61 =0.24
737
HazusMHTechnicalManual
P[ D
s
= ds
3
] =0.61 - 0.55 =0.06
P[ D
s
= ds
4
] =0.55 - 0.47 =0.08
P[ D
s
= ds
5
] =0.47

These discrete values will then be used in the evalution of functionality and economic
losses.

7.2.9 Guidance for Loss Estimation Using Advanced Data and Models Analysis

For this advanced level of analysis, the expert can take advantage of the methodologys
flexibility to (1) include a more refined inventory of the railway system pertaining to the
area of study, and (2) include component-specific and system-specific fragility data.
Default User-Supplied Data Analysis damage algorithms can be modified, or replaced, to
incorporate improved information about key components of a railway system, such as
urban stations. Similarly, better restoration curves can be developed, given knowledge of
available resources and a more accurate layout of the railway network within the local
topographic and geological conditions (i.e., if the redundancy and importance of railway
components of the network are known).

7.2.10 References

Applied Technology Council, "Earthquake Damage Evaluation Data for California",
ATC-13, Redwood City, CA, 1985.

G&E Engineering Systems, Inc. (G&E), "NIBS Earthquake Loss Estimation Methods,
Technical Manual, Transportation Systems (Railway Systems)", May 1994.



738
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
Time (days)

P
e
r
c
e
n
t

F
u
n
c
t
i
o
n
a
l

0
25
50
75
100
1 10 100 1000
Minor Moderate Extensive Complete ATC-13 Data

Figure 7.11 Restoration Curves for Railway Tracks/Roadbeds.



Time (days)

P
e
r
c
e
n
t

F
u
n
c
t
i
o
n
a
l

0
25
50
75
100
1 10 100 1000
Minor Moderate Extensive Complete ATC-13 Data

Figure 7.12 Restoration Curves for Railway Bridges.


739
HazusMHTechnicalManual
Time (days)

P
e
r
c
e
n
t

F
u
n
c
t
i
o
n
a
l

0
25
50
75
100
1 10 100 1000
Minor Moderate Extensive Complete ATC-13 Data

Figure 7.13 Restoration Curves for Railway Tunnels.


Time (days)

P
e
r
c
e
n
t

F
u
n
c
t
i
o
n
a
l

0
25
50
75
100
1 10 100 1000
Minor Moderate Extensive Complete ATC-13 Data

Figure 7.14 Restoration Curves for Railway Facilities.

740
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
Peak Ground Acceleration (g)
P
r
o
b
a
b
i
l
i
t
y

[

D
s


>


d
s


|


P
G
A

]







0.0000
0.2500
0.5000
0.7500
1.0000
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Slight/Minor Moderate Extensive Complete

Figure 7.15 Fragility Curves at Various Damage States for Seismically Designed
Railway Bridges Subject to Peak Ground Acceleration.

Peak Ground Acceleration (g)
P
r
o
b
a
b
i
l
i
t
y

[

D
s


>


d
s


|


P
G
A

]







0.0000
0.2500
0.5000
0.7500
1.0000
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Slight/Minor Moderate Extensive Complete

Figure 7.16 Fragility Curves at Various Damage States for Conventionally Designed
Railway Bridges Subject to Peak Ground Acceleration.




741
HazusMHTechnicalManual
Permanent Ground Deformation (inches)
P
r
o
b
a
b
i
l
i
t
y

[

D
s


>


d
s


|


P
G
D

]






0
0.25
0.5
0.75
1
0.00 6.00 12.00 18.00 24.00 30.00
Slight/Minor Moderate Extensive Complete

Figure 7.17 Fragility Curves at Various Damage States for Seismically-Designed
Railway Bridges Subject to Permanent Ground Deformation.


Permanent Ground Deformation (inches)
P
r
o
b
a
b
i
l
i
t
y

[

D
s


>


d
s


|


P
G
D

]






0
0.25
0.5
0.75
1
0.00 6.00 12.00 18.00 24.00 30.00
Slight/Minor Moderate Extensive Complete

Figure 7.18 Fragility Curves at Various Damage States for Conventionally-Designed
Railway Bridges Subject to PGD.






742
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems

Top event occurs if any
of the bottom events occur
Sub-
Component
Fuel Facility
Component
Equipment
Pump Building Tanks
Top event occurs if all
of the bottom events occur
Off-Site Power Backup Power


Figure 7.19a Fault Tree for Moderate Damage to Fuel Facilities with Anchored
Equipment and Backup Power.



Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y

D
s


>


d
s


|


P
G
A

]


0.00
0.25
0.50
0.75
1.00
0.00 0.20 0.40 0.60 0.80 1.00 1.20
Off-Site Power
Tank
Backup Power
Fragility Curve For a
Fuel Facility
(Component)
Subcomponents
Equipment
Pump Building


Figure 7.19b An Example of Fitting a Lognormal Curve (solid line) to a Fuel
Facility Fragility Curve (dotted line).




743
HazusMHTechnicalManual
Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y

D
s


>


d
s


|


P
G
A

]




0.00
0.25
0.50
0.75
1.00
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive Complete

Figure 7.20.a Fragility Curves at Various Damage States for Fuel Facility with
Anchored Components and Backup Power.


Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y

D
s


>


d
s


|


P
G
A

]





0.00
0.25
0.50
0.75
1.00
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive Complete

Figure 7.20.b Fragility Curves at Various Damage States for Fuel Facility with
Anchored Components but no Backup Power.


744
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y

D
s


>


d
s


|


P
G
A

]





0.00
0.25
0.50
0.75
1.00
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive Complete

Figure 7.20.c Fragility Curves at Various Damage States for Fuel Facility with
Unanchored Components and Backup Power.

Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y

D
s


>


d
s


|


P
G
A

]





0.00
0.25
0.50
0.75
1.00
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive Complete

Figure 7.20.d Fragility Curves at Various Damage States for Fuel Facility with
Unanchored Components and no Backup Power.



745
HazusMHTechnicalManual










Permanent Ground Deformation (inches)
[

P
r
o
b
a
b
i
l
i
t
y


D
s


>


d
s


|


P
G
D

]







0
0.25
0.5
0.75
1
0.00 6.00 12.00 18.00 24.00 30.00 36.00 42.00 48.00
Minor Moderate Extensive/Complete

Figure 7.20.e Fragility Curves at Various Damage States for Fuel Facility with
Buried Tanks Subject to Permanent Fround Deformation.











746
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y


D
s

>

d
s

|


P
G
A

]



0.000
0.250
0.500
0.750
1.000
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive Complete

Figure 7.21.a Fragility Curves at Various Damage States for Dispatch Facility with
Anchored Components and Backup Power.


Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y


D
s

>

d
s

|


P
G
A

]



0.000
0.250
0.500
0.750
1.000
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive Complete

Figure 7.21.b Fragility Curves at Various Damage States for Dispatch Facility with
Anchored Components but no Backup Power.




747
HazusMHTechnicalManual
Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y


D
s

>

d
s

|


P
G
A

]



0.000
0.250
0.500
0.750
1.000
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive Complete

Figure 7.21.c Fragility Curves at Various Damage States for Dispatch Facility with
Unanchored Components and Backup Power.


Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y


D
s

>

d
s

|


P
G
A

]



0.000
0.250
0.500
0.750
1.000
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive Complete

Figure 7.21.d Fragility Curves at Various Damage States for Dispatch Facility with
Unanchored Components and no Backup Power.


748
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
7.3 Light Rail Transportation System

7.3.1 Introduction

This section presents an earthquake loss estimation methodology for a light rail
transportation system. Like railway systems, light rail systems consist of railway
tracks/roadbeds, bridges, tunnels, maintenance facilities, dispatch facilities and DC power
substations. Therefore, the only difference in the case of light rail systems is in the fuel
facilities, which are DC power substations.

7.3.2 Scope

The scope of this section includes development of methods for estimation of earthquake
damage to a light rail transportation system given knowledge of the system's components,
the classification of each component (e.g., for dispatch facilities, whether the facility's
equipment is anchored or not), and the ground motion (i.e. peak ground acceleration
and/or permanent ground deformation).

Damage states describing the level of damage to each light rail system component are
defined (i.e. slight, moderate, extensive or complete). Damage states are related to
damage ratio (defined as ratio of repair to replacement cost) for evaluation of direct
economic loss. Component restoration curves are provided for each damage state to
evaluate loss of function.

Fragility curves are developed for each type of light rail system component. These
curves describe the probability of reaching or exceeding each damage state given the
level of ground motion.

Interdependence of components on overall system functionality is not addressed by the
methodology. Such considerations require a system (network) analysis that would be
performed separately by a light rail system expert as an advanced study.

7.3.3 Input Requirements and Output Information

Required input to estimate damage to light rail systems includes the following items:

Light Rail Tracks/Roadbeds

- Geographical location of railway links [longitude and latitude of end nodes]
- Permanent ground deformation (PGD) at roadbed link

Light Rail Bridges

- Geographical location of bridge [longitude and latitude]
- Spectral values and PGD at bridge
- Bridge classification
749
HazusMHTechnicalManual

Light Rail Tunnels

- Geographical location of tunnels [longitude and latitude]
- PGA and PGD at tunnel
- Tunnel Classification

Light Rail Facilities (DC substations, maintenance and dispatch facilities)

- Geographical location of facilities [longitude and latitude]
- PGA and PGD at facility
- Classification

Direct damage output for light rail systems includes probability estimates of (1)
component functionality and (2) physical damage, expressed in terms of the component's
damage ratio. Note that damage ratios, which are the inputs to direct economic loss
methods, are described in section 15.3 of Chapter 15.

Component functionality is described by the probability of being in a damage state
(immediately following the earthquake) and by the associated fraction or percentage of
the component that is expected to be functional after a specified period of time.

7.3.4 Form of Damage Functions

Damage functions or fragility curves for all light rail system components mentioned
above are modeled as lognormal functions that give the probability of reaching or
exceeding different levels of damage for a given level of ground motion. Each fragility
curve is characterized by a median value of ground motion (or failure) and an associated
dispersion factor (lognormal standard deviation). Ground motion is quantified in terms
of peak ground acceleration (PGA) and ground failure is quantified in terms of permanent
ground displacement (PGD).

- Fragility curves for tracks/roadbeds are the same as for railway tracks/roadbeds.
- Fragility curves for bridges are the same as for railway bridges.
- Fragility curves for tunnels are the same as for railway tunnels.
- Fragility curves for maintenance and dispatch facilities are the same as for railway
maintenance and dispatch facilities.
- Fragility curves for DC power substations are defined in terms of PGA and PGD.

7.3.5 Description of Light Railway System Components

A light rail system consists mainly of six components: tracks/roadbeds, bridges, tunnels,
maintenance facilities, dispatch facilities, and DC power substations. The first five are
the same as for railway systems and are already described in Section 7.2. Therefore, only
DC substations will be described in this subsection.

750
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
DC Power Substations

Light rail systems use electric power and have low voltage DC power substations.
DC power is used by the light rail system's electrical distribution system. The DC
power substations consist of electrical equipment, which convert the local electric
utility AC power to DC power. Two types of DC power stations are considered.
These are: (1) DC power stations with anchored (seismically designed) components
and (2) DC power stations with unanchored (which are not seismically designed)
components.

7.3.6 Definitions of Damage States

A total of five damage states are defined for light rail system components. These are
none (ds
1
), slight/minor (ds
2
), moderate (ds
3
), extensive (ds
4
) and complete (ds
5
).


Slight or Minor Damage (ds
2
)

- For tracks/roadbeds, ds
2
is defined similar to railway tracks.

- For light rail bridges, ds
2
is defined similar to railway bridges.

- For light rail tunnels, ds
2
is defined similar to highway tunnels.

- For light rail system facilities,

^ For maintenance facilities, ds
2
is defined similar to railway maintenance
facilities.

^ For dispatch facilities, ds
2
is defined similar to railway dispatch facilities.

^ For DC power substations with anchored or unanchored components, ds
2
is
defined by loss of off-site power for a very short period, or slight damage to
building.

Moderate Damage (ds
3
)

- For tracks/roadbeds, ds
3
is defined similar to railway tracks.

- For light rail bridges, ds
3
is defined similar to railway bridges.

- For light rail tunnels, ds
3
is defined similar to highway tunnels.

- For light rail system facilities,
751
HazusMHTechnicalManual

^ For maintenance facilities, ds
3
is defined similar to railway maintenance
facilities.

^ For dispatch facilities, ds
3
is defined similar to railway dispatch facilities.

^ For DC power substations with anchored or unanchored components, ds
3
is
defined by loss of off-site power for few days, considerable damage to equipment,
or moderate damage to building.

Extensive Damage (ds
4
)

- For tracks/roadbeds, ds
4
is defined similar to railway tracks.

- For light rail bridges, ds
4
is defined similar to railway bridges.

- For light rail tunnels, ds
4
is defined similar to highway tunnels.

- For light rail system facilities,

^ For maintenance facilities, ds
4
is defined similar to railway maintenance
facilities.

^ For dispatch facilities, ds
4
is defined similar to railway dispatch facilities.

^ For DC power substations with anchored or unanchored components, ds
4
is
defined by extensive building damage.

Complete Damage (ds
5
)

- For tracks/roadbeds, ds
5
is defined similar to railway tracks.

- For light rail bridges, ds
5
is defined similar to railway bridges.

- For light rail tunnels, ds
5
is defined similar to highway tunnels.

- For light rail system facilities,

^ For maintenance facilities, ds
5
is defined similar to railway maintenance
facilities.

^ For dispatch facilities, ds
5
is defined similar to railway dispatch facilities.

752
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
^ For DC power substations with anchored or unanchored components, ds
5
is
defined by complete building damage.

7.3.7 Component Restoration Curves

The restoration curves for light rail tracks/roadbeds, bridges, tunnels, and facilities are
assumed to be the same as those for railway system components.

7.3.8 Development of Damage Functions

Fragility curves for light rail system components are defined with respect to classification
and ground motion parameter. Again, except for DC power stations, damage functions of
the other light rail system components have been already established in either section 7.1
(highway systems) or section 7.2 (railway systems).

Damage functions for Light Rail Tracks/Roadbeds

See damage functions for railway tracks/roadbeds.

Damage Functions for Light Rail Bridges

See damage functions for railway bridges.

Damage Functions for Light Rail Tunnels

See damage functions for highway tunnels.

Damage Functions for Light Rail System Facilities

Damage functions for light rail system facilities are defined in terms of PGA and
PGD. Note that ground failure (PGD) related damage functions for these
facilities are assumed to be similar to those described for railway system facilities
in section 7.2.8.

PGA Related Damage Functions for Maintenance Facilities

Maintenance facilities for light rail systems are mostly made of steel braced frames.
Since no default inventory is provided for these facilities, the user will be expected to
provide the appropriate mapping between these facilities whose damage functions are
listed in Table 7.7 of section 7.2.8 and their model building types.

PGA Related Damage Functions for Dispatch Facilities

See damage functions for railway dispatch facilities.

PGA Related Damage Functions for DC Power Substations
753
HazusMHTechnicalManual
Fragility curves for the two types of DC power substations are developed based on
the type of damage incurred by the DC power substation subcomponents (building,
equipment, and off-site power for interaction effects). These two types are DC power
substations with unanchored equipment, and DC power substations with unanchored
equipment. Medians and dispersions of damage functions to DC power substations
subcomponents are summarized in Tables C.7.1 and C.7.2 of Appendix 7C.
Component fragility curves are obtained using the same methodology as used before.
That is, each fragility curve is determined by a lognormal curve that best fits the
results of the Boolean combination. It should be mentioned that the Boolean logic is
implicitly presented within the definition of a particular damage state. The medians
and dispersions of the damage functions for anchored and unanchored DC power
substations are shown in Table 7.13 and plotted in Figures 7.22.a and 7.22.b.

Table 7.13 Damage Algorithms for DC Power Substations
Peak Ground Acceleration
Classification Damage State Median (g) |
Substation with
Anchored
Components
slight
moderate
extensive
complete
0.12
0.27
0.80
1.50
0.55
0.45
0.80
0.80
Substation with
Unnchored
Components
slight
moderate
extensive
complete
0.11
0.23
0.80
1.50
0.50
0.40
0.80
0.80

7.3.9 Guidance for Loss Estimation Using Advanced Data and Models Analysis

For this type of analysis, the expert can use the methodology developed with the
flexibility to (1) include a refined inventory of the light rail system pertaining to the area
of study, and (2) include component specific and system specific fragility data. Default
User-Supplied Data Analysis damage algorithms can be modified or replaced to
accommodate any specified key component of a light railway system, such as a bridge.
Similarly, better restoration curves could be developed given knowledge of available
resources and a more accurate layout of the light rail network within the local
topographic and geological conditions (i.e. redundancy and importance of a light railway
component in the network are known).

7.3.10 References

Applied Technology Council, "Earthquake Damage Evaluation Data for California",
ATC-13, Redwood City, CA, 1985.

G & E Engineering Systems, Inc. (G & E), "NIBS Earthquake Loss Estimation Methods,
Technical Manual, Transportation Systems", May 1994.

754
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y


D
s

>

d
s

|


P
G
A

]



0.000
0.250
0.500
0.750
1.000
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive Complete

Figure 7.22.a Fragility Curves at Various Damage States for DC Power Substations
with Anchored Components.


Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y


D
s

>

d
s

|


P
G
A

]



0.000
0.250
0.500
0.750
1.000
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive Complete

Figure 7.22.b Fragility Curves at Various Damage States for DC Power Substations
with Unanchored Components.

755
HazusMHTechnicalManual
7.4 Bus Transportation System

7.4.1 Introduction

This section presents a loss estimation methodology for a bus transportation system
during earthquakes. Bus facilities consist of maintenance, fuel, and dispatch facilities.
The facilities may sustain damage due to ground shaking or ground failure. Major losses
can occur if bus maintenance buildings collapse, and operational problems may arise if a
dispatch facility is damaged.

7.4.2 Scope

The scope of this section includes development of methods for estimation of earthquake
damage to a bus transportation system given knowledge of components (i.e., fuel,
maintenance, and dispatch facilities with or without backup power), classification (i.e. for
fuel facilities, anchored or unanchored components), and the ground motion (i.e. peak
ground acceleration and/or permanent ground deformation).

Damage states describing the level of damage to each of the bus system components are
defined (i.e. slight, moderate, extensive or complete). Damage states are related to
damage ratio (defined as ratio of repair to replacement cost) for evaluation of direct
economic loss. Component restoration curves are provided for each damage state to
evaluate loss of function. Restoration curves describe the fraction or percentage of the
component that is expected to be open or operational as a function of time following the
earthquake. For bus systems, the restoration is dependent upon the extent of damage to
the fuel, maintenance, and dispatch facilities.

Fragility curves are developed for each class of bus system component. These curves
describe the probability of reaching or exceeding each damage state given the level of
ground motion. Based on these fragility curves, a method for assessing functionality of
each of the three bus system components is presented.

Interdependence of components on overall system functionality is not addressed by the
methodology. Such considerations require a system (network) analysis that would be
performed separately by a bus system expert as an advanced study.

7.4.3 Input Requirements and Output Information

Required input to estimate damage to bus systems includes the following items:

Urban Stations

- Geographical location of site
- Spectral values and PGD at station
- Classification

756
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
Fuel Facilities

- Geographical location of site
- PGA and PGD at facility
- Classification (i.e. with or without anchored equipment and backup power)

Maintenance Facilities

- Geographical location of site
- Spectral values and PGD at facility
- Classification (i.e. building type)

Dispatch Facilities

- Geographical location of each warehouse
- PGA and PGD at facility
- Classification (i.e. with or without anchored equipment and backup power)

Direct damage output for bus systems includes probability estimates of (1) component
functionality and (2) physical damage, expressed in terms of the component's damage
ratio.

Component functionality is described by the probability of being in a damage state
(immediately following the earthquake) and by the associated fraction or percentage of
the component that is expected to be functional after a specified period of time.

7.4.4 Form of Damage Functions

Damage functions or fragility curves for all three bus system components, mentioned
above, are lognormal functions that give the probability of reaching or exceeding
different levels of damage for a given level of ground motion. Each fragility curve is
characterized by a median value of ground motion (or failure) and an associated
dispersion factor (lognormal standard deviation). Ground motion is quantified in terms
of peak ground acceleration (PGA) and ground failure is quantified in terms of permanent
ground displacement (PGD).
- For urban stations, the fragility curves are defined in terms of Spectral values and
PGD.

- For fuel facilities, the fragility curves are defined in terms of PGA and PGD.

- For maintenance facilities, the fragility curves are defined in terms of Spectral values
and PGD.

- For dispatch facilities, the fragility curves are defined in terms of PGA and PGD.
Definitions of various damage states and the methodology used in deriving all these
fragility curves are presented in the following section.
757
HazusMHTechnicalManual

7.4.5 Description of Bus System Components

A bus system consists mainly of four components: urban stations, fuel facilities,
maintenance facilities, and dispatch facilities. This section provides a brief description of
each.

Urban Stations

These are mainly buildings structures.

Bus System Fuel Facilities

Fuel facility consists of fuel storage tanks, buildings, pump equipment and buried
pipe, and, sometimes, backup power. The fuel facility functionality is determined
with a fault tree analysis considering redundancies and sub-component behavior. The
same classes assumed for railway fuel facilities are assumed here. These are listed in
Table 3.9.

Bus System Maintenance Facilities

Maintenance facilities for bus systems are mostly made of steel braced frames. The
same classes assumed for railway maintenance facilities are assumed here.

Bus System Dispatch Facilities

The same classes assumed for railway dispatch facilities are assumed here. These
are listed in Table 3.9.

7.4.6 Definitions of Damage States

A total of five damage states are defined for highway system components. These are
none (ds
1
), slight/minor (ds
2
), moderate (ds
3
), extensive (ds
4
) and complete (ds
5
).

Slight Damage (ds
2
)

^ For urban stations, ds
2
is defined similar to railway urban stations.

^ For fuel facilities, ds
2
is defined similar to railway fuel facilities.

^ For maintenance facilities, ds
2
is defined similar to railway maintenance
facilities.

^ For dispatch facilities, ds
2
is defined similar to railway dispatch facilities.

758
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
Moderate Damage (ds
3
)

^ For urban stations, ds
3
is defined similar to railway urban stations.

^ For fuel facilities, ds
3
is defined similar to railway fuel facilities.

^ For maintenance facilities, ds
3
is defined similar to railway maintenance
facilities.

^ For dispatch facilities, ds
3
is defined similar to railway dispatch facilities.

Extensive Damage (ds
4
)

^ For urban stations, ds
4
is defined similar to railway urban stations.

^ For fuel facilities, ds
4
is defined similar to railway fuel facilities.

^ For maintenance facilities, ds
4
is defined similar to railway maintenance
facilities.

^ For dispatch facilities, ds
4
is defined similar to railway dispatch facilities.

Complete Damage (ds
5
)

^ For urban stations, ds
5
is defined similar to railway urban stations.

^ For fuel facilities, ds
5
is defined similar to railway fuel facilities.

^ For maintenance facilities, ds
5
is defined similar to railway maintenance
facilities.

^ For dispatch facilities, ds
5
is defined similar to railway dispatch facilities.

7.4.7 Component Restoration Curves

Restoration Curves are developed based on a best fit to ATC-13 damage data for the
social functions SF 26a through SF 26d, consistent with damage states defined in the
previous section. Normal distribution functions are developed using the ATC-13 data for
the mean time for 30%, 60% and 100% restoration of different sub-components in
different damage states. The restoration curves for bus transportation systems are similar
to those of railway transportation systems. Means and dispersions of these restoration
functions are given in Tables 7.10.a. Discretized restoration functions are shown in Table
7.10.b, where the percentage restoration is shown at discrete times.
759
HazusMHTechnicalManual

7.4.8 Development of Damage Functions

Fragility curves for bus system components are defined with respect to classification and
ground motion parameter.

Damage Functions for Bus System Urban Stations

Urban stations are classified based on the building structural type. Damage functions
for bus system urban stations are similar to those for the railway transportation
system (see Section 7.2.8).

Damage Functions for Bus System Fuel Facilities

Fuel facilities are classified based on two criteria: (1) whether the sub-components
comprising the fuel facilities are anchored or unanchored and (2) whether backup
power exists in the facility. Damage functions for bus system fuel facilities are
similar to those for the railway transportation system (see Section 7.2.8).

Damage Functions for Bus System Maintenance Facilities

The PGA and PGD median values for the damage states of maintenance facilities are
similar to those of light rail maintenance facilities presented in Section 7.3.8.

Damage Functions for Bus System Dispatch Facility

The PGA and PGD median values for the damage states of dispatch facilities are
similar to those of railway dispatch facilities given in Section 7.2.8.

7.4.9 Guidance for Loss Estimation using Advanced Data and Models Analysis

For this level of analysis, the expert can use the methodology developed with the
flexibility to: (1) include a refined inventory of the bus system pertaining to the area of
study, and (2) include component specific and system specific fragility data. Default
User-Supplied Data Analysis damage algorithms can be modified or replaced to
accommodate any specified key component of a bus system, such as a warehouse.
Similarly, better restoration curves could be developed given knowledge of available
resources and a more accurate layout of the bus transportation network within the local
topographic and geological conditions (i.e., redundancy and importance of a bus system
component in the network are known).






760
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
7.4.10 References

Applied Technology Council, "Earthquake Damage Evaluation Data for California",
ATC-13, Redwood City, CA, 1985.

G & E Engineering Systems, Inc. (G & E), "NIBS Earthquake Loss Estimation Methods,
Technical Manual, Transportation Systems", May 1994.

761
HazusMHTechnicalManual
7.5 Port Transportation System

7.5.1 Introduction

This section presents a loss estimation methodology for a port transportation system.
Port facilities consist of waterfront structures (e.g., wharfs, piers and seawalls); cranes
and cargo handling equipment; fuel facilities; and warehouses. In many cases, these
facilities were constructed prior to widespread use of engineered fills; consequently, the
wharf, pier, and seawall structures are prone to damage due to soil failures such as
liquefaction. Other components may be damaged due to ground shaking as well as
ground failure.

7.5.2 Scope

The scope of this section includes developing methods for estimating earthquake damage
to a port transportation system given knowledge of components (i.e., waterfront
structures, cranes and cargo handling equipment, fuel facilities, and warehouses),
classification (i.e. for fuel facilities, anchored or unanchored components, with or without
back-up power), and the ground motion (i.e. peak ground acceleration and/or permanent
ground deformation).

Damage states describing the level of damage to each of the port system components are
defined (i.e. slight, moderate, extensive or complete). Damage states are related to
damage ratio (defined as ratio of repair to replacement cost) for evaluation of direct
economic loss. Component restoration curves are provided for each damage state to
evaluate loss of function. Restoration curves describe the fraction or percentage of the
component that is expected to be open or operational as a function of time following the
earthquake. For ports the restoration is dependent upon the extent of damage to the
waterfront structures, cranes/cargo handling equipment, fuel facilities, and warehouses.
From the standpoint of functionality of the port, the user should consider the restoration
of only the waterfront structures and cranes since the fuel facilities and warehouses are
not as critical to the functionality of the port.

Fragility curves are developed for each class of port system component. These curves
describe the probability of reaching or exceeding a certain damage state given the level of
ground motion. Based on these fragility curves, a method for assessing functionality of
each of the four port system components is presented.

Interdependence of components on overall system functionality is not addressed by the
methodology. Such considerations require a system (network) analysis that would be
performed separately by a port system expert as an advanced study.

7.5.3 Input Requirements and Output Information

Required input to estimate damage to port systems includes the following items:

762
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
Waterfront Structures

- Geographic location of port (longitude and latitude)
- PGA & PGD
- Classification

Cranes/Cargo Handling Equipment

- Geographic location of port (longitude and latitude)
- PGA and PGD
- Classification (i.e. stationary or rail mounted)

Fuel Facilities

- Geographical location of facility [longitude and latitude]
- PGA and PGD
- Classification

Warehouses

- Geographical location of warehouse [longitude and latitude]
- PGA and PGD
- Classification (i.e. building type)

Direct damage output for port systems includes probability estimates of (1) component
functionality and (2) physical damage, expressed in terms of the component's damage
ratio. Damage ratios are used as inputs to direct economic loss methods, as described in
section 15.3 of Chapter 15.

7.5.4 Form of Damage Functions

Damage functions or fragility curves for all four port system components, mentioned
above, are lognormally distributed functions that give the probability of reaching or
exceeding different levels of damage for a given level of ground motion. Each fragility
curve is characterized by a median value of ground motion (or failure) and an associated
dispersion factor (lognormal standard deviation). Ground motion is quantified in terms
of peak ground acceleration (PGA) and ground failure is quantified in terms of permanent
ground displacement (PGD).

- For waterfront structures, the fragility curves are defined in terms of PGD and PGA.

- For cranes/cargo handling equipment, the fragility curves are defined in terms of
PGA and PGD.

- For fuel facilities, the fragility curves are defined in terms of PGA and PGD.

763
HazusMHTechnicalManual
- For warehouses, the fragility curves are defined in terms of PGA and PGD.

Definitions of various damage states and the methodology used in deriving all these
fragility curves are presented in the following section.

7.5.5 Description of Port Components

A port system consists of four components: waterfront structures, cranes/cargo handling
equipment, fuel facilities, and warehouses. This section provides a brief description of
each.

Waterfront Structures

This component includes wharves (port embankments), seawalls (protective walls
from erosion), and piers (break-water structures which form harbors) that exist in the
port system. Waterfront structures typically are supported by wood, steel or concrete
piles. Many also have batter piles to resist lateral loads from wave action and impact
of vessels. Seawalls are caisson walls retaining earth fill material.

Cranes and Cargo Handling Equipment

These are large equipment items used to load and unload freight from vessels. These
are can be stationary or mounted on rails.

Port Fuel Facilities

The fuel facility consists mainly of fuel storage tanks, buildings, pump equipment,
piping, and, sometimes, backup power. These are the same as those for railway
systems presented in Section 7.2. The functionality of fuel systems is determined
with a fault tree analysis, which considers redundancies and sub-component behavior,
as it can be seen in Figures 7.18 and 7.19 of Section 7.2. Note that five types of fuel
facilities in total are defined.

Warehouses

Warehouses are large buildings usually constructed of structural steel. In some cases,
warehouses may be several hundred feet from the shoreline, while in other instances;
they may be located on the wharf itself.

7.5.6 Definition of Damage States

A total of five damage states are defined for port system components. These are none
(ds
1
), slight/minor (ds
2
), moderate (ds
3
), extensive (ds
4
) and complete (ds
5
).

Slight/Minor Damage (ds
2
)

764
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
- For waterfront structures, ds
2
is defined by minor ground settlement resulting in
few piles (for piers/seawalls) getting broken and damaged. Cracks are formed on
the surface of the wharf. Repair may be needed.

- For cranes/cargo handling equipment, ds
2
is defined by slight damage to structural
members with no loss of function for the stationary equipment, while for the
unanchored or rail mounted equipment, ds
1
is defined as minor derailment or
misalignment without any major structural damage to the rail mount. Minor repair
and adjustments may be required before the crane becomes operable.

- For fuel facilities, ds
2
is defined the same as for railway facilities.

- For warehouses, ds
2
is defined by slight damage to the warehouse building.

Moderate Damage (ds
3
)

- For waterfront structures, ds
3
is defined as considerable ground settlement with
several piles (for piers/seawalls) getting broken and damaged.

- For cranes/cargo handling equipment, ds
3
is defined as derailment due to
differential displacement of parallel track. Rail repair and some repair to
structural members is required.

- For fuel facilities, ds
3
is defined the same as for railway facilities.

- For warehouses, ds
3
is defined by moderate damage to the warehouse building.

Extensive Damage (ds
4
)

- For waterfront structures, ds
4
is defined by failure of many piles, extensive sliding
of piers, and significant ground settlement causing extensive cracking of
pavements.

- For cranes/cargo handling equipment, ds
4
is defined by considerabe damage to
equipment. Toppled or totally derailed cranes are likely to occcur. Replacement
of structural members is required.

- For fuel facilities, ds
4
is defined same as for railway facilities.

- For warehouses, ds
4
is defined by extensive damage to warehouse building.

Complete Damage (ds
5
)

- For waterfront structures, ds
5
is defined as failure of most piles due to significant
ground settlement. Extensive damage is widespread at the port facility.
765
HazusMHTechnicalManual

- For cranes/cargo handling equipment, ds
5
is the same as ds
4
.

- For fuel facilities with buried tanks, ds
5
is the same as for railway facilities.

- For warehouses, ds
5
is defined by total damage to the warehouse building.

7.5.7 Component Restoration Curves

Restoration Curves are developed based on a best fit to ATC-13 damage data for social
functions SF 28.a and SF 29.b, consistent with damage states defined in the previous
section. Normal distribution functions are developed using the ATC-13 data for the mean
time for 30%, 60% and 100% restoration of different sub-components in different
damage states. Means and dispersions of these restoration functions are given in Table
7.14.a. The discretized restoration functions are given in Table 7.14.b, where the
percentage restoration is shown at some specified time intervals. These restoration
functions are shown in Figures 7.23 and 7.24. Figure 7.23 represents restoration curves
for waterfront structures, while Figure 7.24 shows restorations curve for cranes and cargo
handling equipment.

Table 7.14.a Restoration Functions for Port Sub-Components
Restoration Functions (All Normal Distributions)
Classification Damage State Mean (Days) o
Buildings,
Waterfront
Structures
slight/minor
moderate
extensive
complete
0.6
3.5
22
85
0.2
3.5
22
73
Cranes/Cargo
Handling
Equipment
slight/minor
moderate
extensive
complete
0.4
6
30
75
0.35
6
30
55

Table 7.14.b Discretized Restoration Functions for Port Sub-Components
Discretized Restoration Functions
Classification Damage State 1 day 3 days 7 days 30 days 90 days
Buildings,
Waterfront
Structures
slight/minor
moderate
extensive
complete
96
24
17
12
100
43
19
13
100
84
25
14
100
100
63
22
100
100
100
53
Cranes/Cargo
Handling
Equipment
slight/minor
moderate
extensive
complete
96
20
17
9
100
31
18
10
100
57
22
11
100
100
50
21
100
100
100
62

766
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
7.5.8 Development of Damage Functions

Damage functions for port system facilities are defined in terms of PGA and PGD. Note
that, unless it is specified otherwise, ground failure (PGD) related damage functions for
these facilities are assumed to be similar to those described for railroad system facilities
in section 7.2.8.

An example of how to combine PGD and PGA algorithms is presented in section 7.2.8.

Damage functions for Waterfront Structures

Damage functions for waterfront structures were established based on damagability of
subcomponents, namely, piers, seawalls, and wharf. Fault tree logic and the
lognormal best fitting technique were used in developing these fragility curves. The
fault tree is implicitly described in the description of the damage state. The obtained
damage functions are shown in Figure 7.25. Their medians and dispersions are
presented in Table 7.15a. Subcomponent damage functions are given in Table 7.D.1
of Appendix 7D.

Table 7.15.a Damage Algorithms for Waterfront Structures
Permanent Ground Deformation
Components Damage State Median (in) |
slight/minor 5 0.50
Waterfront moderate 12 0.50
Structures extensive 17 0.50
(PWS1) complete 43 0.50

Damage Functions for Cranes and Cargo Handling Equipment

For cranes, a distinction is made between stationary and rail-mounted cranes. The
medians and dispersions of damage functions are presented in Tables 7.15.b, while the
fragility curves are shown in Figures 7.26 through 7.29.

767
HazusMHTechnicalManual
Table 7.15.b Damage Algorithms for Cranes/Cargo Handling Equipment

Peak Ground Acceleration
Classification Damage State Median (g) |
Anchored/
Stationary
(PEQ1)
slight/minor
moderate
extensive/complete
0.3
0.5
1.0
0.6
0.6
0.7
Unanchored/Rail
mounted (PEQ2)
slight/minor
moderate
extensive/complete
0.15
0.35
0.8
0.6
0.6
0.7

Permanent Ground Deformation
Classification Damage State Median (in) |
Anchored/
Stationary
(PEQ1)
slight/minor
moderate
extensive/complete
3
6
12.0
0.6
0.7
0.7
Unanchored/Rail
mounted (PEQ2)
slight/minor
moderate
extensive/complete
2
4.0
10
0.6
0.6
0.7

Damage Functions for Port System Fuel Facilities

Damage funcitons for fuel facilities are similar to those developed for railway fuel
facilities in Section 7.2.8.


PGA Related Damage Functions for Warehouses

Since no default inventory is provided for these facilities, the user will be
expected to provide the appropriate mapping bewteen these facilities and the
building types which are assumed to be the same as for railway maintenance
facilities whose damage functions are listed in Table 7.7 of section 7.2.8.

7.5.9 Guidance for Loss Estimation using Advanced Data and Models Analysis

For this type of analysis, the expert can use the methodology developed with the
flexibility to: (1) include a refined inventory of the port transportation system pertaining
to the area of study, and (2) include component specific and system specific fragility data.
Default User-Supplied Data Analysis damage algorithms can be modified or replaced to
accommodate any specified key component of a port system, such as a warehouse.
Similarly, better restoration curves could be developed given knowledge of available
resources and a more accurate layout of the port network within the local topographic and
geological conditions (i.e., redundancy and importance of a port system component in the
network are known).

768
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
7.5.10 References

Applied Technology Council, "Earthquake Damage Evaluation Data for California",
ATC-13, Redwood City, CA, 1985.

G & E Engineering Systems, Inc. (G & E), "NIBS Earthquake Loss Estimation Methods,
Technical Manual, Transportation Systems", May 1994.

769
HazusMHTechnicalManual
Time (days)

P
e
r
c
e
n
t

F
u
n
c
t
i
o
n
a
l

0
25
50
75
100
1 10 100 1000
Minor Moderate Extensive Complete ATC-13 Data

Figure 7.23 Restoration Curves for Port Waterfront Structures.


Time (days)

P
e
r
c
e
n
t

F
u
n
c
t
i
o
n
a
l

0
25
50
75
100
1 10 100 1000
Minor Moderate Extensive Complete ATC-13 Data

Figure 7.24 Restoration Curves for Cranes/Cargo Handling Equipment.






770
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems

Permanent Ground Deformation (inches)
P
r
o
b
a
b
i
l
i
t
y

[

D
s


>


d
s


|


P
G
D

]






0
0.25
0.5
0.75
1
0.00 6.00 12.00 18.00 24.00 30.00
Slight/Minor Moderate Extensive Complete


Figure 7.25 Fragility Curves for Waterfront Structures.


771
HazusMHTechnicalManual
Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y

D
s


>


d
s


|


P
G
A

]





0.00
0.25
0.50
0.75
1.00
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive/Complete

Figure 7.26 Fragility Curves for Stationary Cranes/Cargo Handling Equipment
Subject to Peak Ground Acceleration.


Permanent Ground Deformation (inches)
[

P
r
o
b
a
b
i
l
i
t
y


D
s


>


d
s


|


P
G
D

]







0
0.25
0.5
0.75
1
0.00 5.00 10.00 15.00 20.00 25.00 30.00
Minor Moderate Extensive/Complete


Figure 7.27 Fragility Curves for Stationary Cranes/Cargo Handling Equipment
Subject to Permanent Ground Deformation.

772
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y

D
s


>


d
s


|


P
G
A

]





0.00
0.25
0.50
0.75
1.00
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive/Complete


Figure 7.28 Fragility Curves for Rail Mounted Cranes/Cargo Handling Equipment
Subject to Peak Ground Acceleration.


Permanent Ground Deformation (inches)
[

P
r
o
b
a
b
i
l
i
t
y


D
s


>


d
s


|


P
G
D

]







0
0.25
0.5
0.75
1
0.00 5.00 10.00 15.00 20.00 25.00 30.00
Minor Moderate Extensive/Complete

Figure 7.29 Fragility Curves for Rail Mounted Cranes/Cargo Handling Equipment
Subject to Permanent Ground Deformation.

773
HazusMHTechnicalManual
7.6 Ferry Transportation System

7.6.1 Introduction

This section presents a loss estimation methodology for a ferry transportation system.
Ferry systems consist of waterfront structures (e.g., wharf, piers and seawalls); fuel,
maintenance, and dispatch facilities; and passenger terminals.

The waterfront structures are located at the points of embarkation or disembarkation, and
they are similar to, although not as extensive as, those of the port transportation system.
In some cases the ferry system may be located within the boundary of the port
transportation system. The points of embarkation or disembarkation are located some
distance apart from one another, usually on opposite shorelines.

Fuel and maintenance facilities are usually located at one of these two points. The size of
the fuel facility is smaller than that of the port facility. In many cases, the dispatch
facility is located in the maintenance facility or one of the passenger terminals.

7.6.2 Scope

The scope of this section includes development of methods for estimation of earthquake
damage to a ferry transportation system given knowledge of components (i.e., waterfront
structures; fuel, maintenance, and dispatch facilities; and passenger terminals),
classification (i.e. for fuel facilities, anchored or unanchored components, with or without
back-up power), and the ground motion (i.e. peak ground acceleration and/or permanent
ground deformation).

Damage states describing the level of damage to each of the ferry system components are
defined (i.e. slight/minor, moderate, extensive or complete). Damage states are related to
damage ratio (defined as ratio of repair to replacement cost) for evaluation of direct
economic loss. Component restoration curves are provided for each damage state to
evaluate loss of function. Restoration curves describe the fraction or percentage of the
component that is expected to be open or operational as a function of time following the
earthquake. For ferries the restoration is dependent upon the extent of damage to the
waterfront structures; fuel, maintenance, and dispatch facilities; and passenger terminals.

Fragility curves are developed for each class of the ferry system components. These
curves describe the probability of reaching or exceeding each damage state given the
level of ground motion. Based on these fragility curves, a method for assessing
functionality of each of the five ferry system components is presented.

Interdependence of components on overall system functionality is not addressed by the
methodology. Such considerations require a system (network) analysis that would be
performed separately by a transportation system expert as an advanced study.

774
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
7.6.3 Input Requirements and Output Information

Required input to estimate damage to ferry systems includes the following items:

Ferry Waterfront Structures
- Geographic locations of harbor
- PGA & PGD

Ferry Fuel Facilities
- Geographical location of facility
- PGA and PGD
- Classification

Ferry Maintenance Facilities
- Geographical location of facility
- Spectral values and PGD
- Classification (i.e. building type)

Ferry Dispatch Facilities
- Geographical location of facility
- PGA and PGD
- Classification

Ferry Terminal Buildings
- Geographical location of building
- Spectral values and PGD
- Classification (i.e. building type)

Direct damage output for ferry systems includes probability estimates of (1) component
functionality and (2) physical damage, expressed in terms of the component's damage
ratio. Damage ratios are used as inputs to direct economic loss methods, as described in
section 15.3 of Chapter 15.

7.6.4 Form of Damage Functions

Damage functions or fragility curves for all five ferry system components mentioned
above, are lognormal functions that give the probability of reaching or exceeding
different levels of damage for a given level of ground motion. Each fragility curve is
characterized by a median value of ground motion (or failure) and an associated
dispersion factor (lognormal standard deviation). Ground motion is quantified in terms
of peak ground acceleration (PGA) and ground failure is quantified in terms of permanent
ground displacement (PGD).




775
HazusMHTechnicalManual

- For waterfront structures, the fragility curves are defined in terms of PGA & PGD.

- For fuel facilities, maintenance and dispatch facilities; and terminal building, the
fragility curves are defined in terms of PGA and PGD.

Definitions of various damage states and the methodology used in deriving fragility
curves for ferry system components are presented in the following subsections.

7.6.5 Description of Ferry System Components

A ferry system consists of the five components mentioned above: waterfront structures,
fuel facilities, maintenance facilities, dispatch facilities, and passenger terminals. This
section provides a brief description of each.

Waterfront Structures
These are the same as those for port systems described in Section 7.5.5.

Fuel Facilities
These facilities are similar to those for port system mentioned in Section 7.5.5.

Maintenance Facilities
These are often steel braced frame structures, but other building types are possible.

Dispatch Facilities
These are similar to those defined for railway system in Section 7.2.5.

Passenger Terminals
These are often moment resisting steel frames, but other building types are possible.

7.6.6 Definitions of Damage States

A total of five damage states are defined for ferry system components. These are none
(ds
1
), slight/minor (ds
2
), moderate (ds
3
), extensive (ds
4
) and complete (ds
5
).

Slight/Minor Damage (ds
2
)

- For waterfront structures, ds
2
is the same as that for waterfront structures in the
port module.

- For fuel facilities, ds
2
is the same as that for fuel facilities in the port module.

- For maintenance facilities, ds
2
is defined by slight damage to building.

776
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
- For dispatch facilities, ds
2
is the same as that for dispatch facilities in the railway
module.

- For passenger terminals, ds
2
is defined by slight damage to building.

Moderate Damage (ds
3
)

- For waterfront structures, ds
3
is the same as that for waterfront structures in the
port module.

- For fuel facilities, ds
3
is the same as that for fuel facilities in the port module.

- For maintenance facilities, ds
3
is defined by moderate damage to building.

- For dispatch facilities, ds
3
is the same as that for dispatch facilities in the railway
module.

- For passenger terminals, ds
3
is defined by moderate damage to building.

Extensive Damage (ds
4
)

- For waterfront structures, ds
4
is the same as that for waterfront structures in the
port module.

- For fuel facilities, ds
4
is the same as that for fuel facilities in the port module.

- For maintenance facilities, ds
4
is defined by extensive damage to building.

- For dispatch facilities, ds
4
is the same as that for dispatch facilities in the railway
module.

- For passenger terminals, ds
4
is defined by extensive damage to building.

Complete Damage (ds
5
)

- For waterfront structures, ds
5
is the same as that for waterfront structures in the
port module.

- For fuel facilities, ds
5
is the same as that for fuel facilities in the port module.

- For maintenance facilities, ds
5
is defined by complete damage to building.

777
HazusMHTechnicalManual
- For dispatch facilities, ds
5
is the same as that for dispatch facilities in the railway
module.

- For passenger terminals, ds
5
is defined as complete damage to building.

778
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
7.6.7 Component Restoration Curves

Ferry systems are made of components that are similar to either those in port systems (i.e.
waterfront structures, fuel facilities), or those in railway systems (i.e. dispatch facilities,
maintenance facilities, passenger terminals). Therefore, restoration curves for ferry
system components can be found in either Section 7.5 or Section 7.2.

7.6.8 Development of Damage Functions

Similar to restoration curves, damage functions for ferry system components can be
found in either Section 7.5 or Section 7.2.

7.6.9 Guidance for Loss Estimation Using Advanced Data and Models Analysis

For this type of analysis, the expert can use the methodology developed with the
flexibility to: (1) include a refined inventory of the ferry system pertaining to the area of
study, and (2) include component specific and system specific fragility data. Default
User-Supplied Data Analysis damage algorithms can be modified or replaced to
accommodate any specified key component of a ferry system, such as a maintenance
facility. Similarly, better restoration curves could be developed given knowledge of
available resources and a more accurate layout of the ferry transportation network within
the local topographic and geological conditions.

7.6.10 References

Applied Technology Council, "Earthquake Damage Evaluation Data for California",
ATC-13, Redwood City, CA, 1985.

G & E Engineering Systems, Inc. (G & E), "NIBS Earthquake Loss Estimation Methods,
Technical Manual, Transportation Systems", May 1994.


779
HazusMHTechnicalManual
7.7 Airport Transportation System

7.7.1 Introduction

This section presents a loss estimation methodology for an airport transportation system.
Airport transportation system consists of runways, control tower, fuel facilities, terminal
buildings, maintenance facilities, hangar facilities, and parking structures. For airports,
control towers are often constructed of reinforced concrete, while terminal buildings and
maintenance facilities are often constructed of structural steel or reinforced concrete.
Fuel facilities are similar to those for railway transportation systems.

7.7.2 Scope

The scope of this section includes development of methods for estimation of earthquake
damage to an airport transportation system given knowledge of components (i.e.
runways, control tower, fuel, and maintenance facilities, terminal buildings, and parking
structures), classification, and ground motion (i.e. peak ground acceleration and/or
permanent ground deformation).

Damage states describing the level of damage to each of the airport system components
are defined (i.e. slight, moderate, extensive or complete). Damage states are related to
damage ratio (defined as ratio of repair to replacement cost) for evaluation of direct
economic loss. Component restoration curves are provided for each damage state to
evaluate loss of function. Restoration curves describe the fraction or percentage of the
component that is expected to be open or operational as a function of time following the
earthquake. For airports, the restoration is dependent upon the extent of damage to the
airport terminals, buildings, storage tanks (for fuel facilities), control tower, and runways.

Fragility curves are developed for each component class of the airport system. These
curves describe the probability of reaching or exceeding each damage state given the
level of ground motion. Based on these fragility curves, a method for assessing
functionality of each of the six airport system components is presented.

7.7.3 Input Requirements and Output Information

Required input to estimate damage to airport systems includes the following items:

Runways

- Geographic location of airport [longitude and latitude]
- PGD

Control Tower

- Geographic location of airport [longitude and latitude]
- PGA and PGD
780
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
- Classification (i.e. building type)

Fuel Facilities

- Geographical location of facility [longitude and latitude]
- PGA and PGD
- Classification

Terminal Buildings

- Geographical location of airport [longitude and latitude]
- Spectral values and PGD
- Classification (i.e. building type)

Maintenance and Hangar Facilities

- Geographical location of facility [longitude and latitude]
- Spectral values and PGD
- Classification (i.e. building type)

Parking Structures

- Geographical location of structure [longitude and latitude]
- Spectral values and PGD
- Classification (i.e. building type)

Direct damage output for airport systems includes probability estimates of (1) component
functionality and (2) physical damage, expressed in terms of the component's damage
ratio. Damage ratios are used as inputs to direct economic loss methods, as described in
section 15.3 of Chapter 15.

7.7.4 Form of Damage Functions

Damage functions or fragility curves for all five airport system components mentioned
above, are lognormal functions that give the probability of reaching or exceeding
different levels of damage for a given level of ground motion. Each fragility curve is
characterized by a median value of ground motion (or failure) and an associated
dispersion factor (lognormal standard deviation). Ground motion is quantified in terms
of peak ground acceleration (PGA) and ground failure is quantified in terms of permanent
ground displacement (PGD).


- For runways, the fragility curves are defined in terms of PGD.

- For control towers, the fragility curves are defined in terms of Spectral values and
PGD.
781
HazusMHTechnicalManual

- For all other facilities, the fragility curves are defined in terms of Spectral values and
PGD.

Definitions of various damage states and the methodology used in deriving these fragility
curves are presented in the following section.

7.7.5 Description of Airport Components

An airport system consists of the six components mentioned above: runways, control
tower, fuel facilities, maintenance facilities, and parking structures. This section provides
a brief description of each.

Runways
This component consists of well-paved "flat and wide surfaces".

Control Tower
Control tower consists of a building and the necessary equipment of air control and
monitoring.

Fuel Facilities
These have been previously defined in Section 7.2.5 of railway systems.

Terminal Buildings
These are similar to urban stations of railway systems from the classification
standpoint (as well as services provided to passengers).

Maintenance Facilities, Hangar Facilities, and Parking Structures
Classification of maintenance facilities is the same as for those in railway systems.
Hangar facilities and parking structures are mainly composed of buildings.

7.7.6 Definitions of Damage States

A total of five damage states are defined for airport system components. These are none
(ds
1
), slight/minor (ds
2
), moderate (ds
3
), extensive (ds
4
) and complete (ds
5
).


Slight/Minor Damage (ds
2
)

- For runways, ds
2
is defined as minor ground settlement or heaving of runway
surface.

- For control tower, ds
2
is defined as slight damage to the building as given in
section 5.3.

- For fuel facilities, ds
2
is the same as that for fuel facilities in the railway module.
782
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems

- For terminal buildings, ds
2
is defined as slight damage to the building as given in
section 5.3.

- For maintenance and hangar facilities, ds
2
is defined as slight damage to the
building as given in section 5.3.

- For parking structures, ds
2
is defined as slight damage to the building as given in
section 5.3.

Moderate Damage (ds
3
)

- For runways, ds
3
is defined same as ds
2
.

- For control tower, ds
3
is defined as moderate damage to the building as given in
section 5.3.

- For fuel facilities, ds
3
is the same as that for fuel facilities in the railway module.

- For terminal buildings, ds
3
is defined as moderate damage to the building as given
in section 5.3.

- For maintenance and hangar facilities, ds
3
is defined as moderate damage to the
building as given in section 5.3.

- For parking structures, ds
3
is defined as moderate damage to the building as given
in section 5.3.

Extensive Damage (ds
4
)

- For runways, ds
4
is defined as considerable ground settlement or considerable
heaving of runway surface.

- For control tower, ds
4
is defined as extensive damage to the building as given in
section 5.3.

- For fuel facilities, ds
4
is the same as that for fuel facilities in the railway module.

- For terminal buildings, ds
4
is defined as extensive damage to the building as given
in section 5.3.

- For maintenance and hangar facilities, ds
4
is defined as extensive damage to the
building as given in section 5.3.

783
HazusMHTechnicalManual
- For parking structures, ds
4
is defined as extensive damage to the building as given
in section 5.3.

Complete Damage (ds
5
)

- For runways, ds
5
is defined as extensive ground settlement or excessive heaving
of runway surface.

- For control tower, ds
5
is defined as complete damage to the building as given in
section 5.3.

- For fuel facilities, ds
5
is the same as that for fuel facilities in the railway module.

- For terminal buildings, ds
5
is defined as complete damage to the building as given
in section 5.3.

- For maintenance and hangar facilities, ds
5
is defined as complete damage to the
building as given in section 5.3.

- For parking structures, ds
5
is defined as complete damage to the building as given
in section 5.3.

7.7.7 Component Restoration Curves

Restoration Curves are developed based on a best fit to ATC-13 data for social functions
SF 27.a and SF 27.b, consistent with damage states defined in the previous section.
Normal distribution functions are developed using this ATC-13 data for the mean time
for 30%, 60% and 100% restoration. Means and dispersions of these restoration
functions are given in Table 7.16.a and shown in Figures 7.30 and 7.31. The discretized
restoration functions are presented in Table 7.16.b, where the percentage restoration is
shown at selected time intervals.

Table 7.16.a Restoration Functions for Airport Components
Restoration Functions (All Normal Distributions)
Classification Damage State Mean (Days) o
Control Towers,
Parking Structures,
Hangar Facilities,
Terminal Building
slight
moderate
extensive
complete
0
1.5
50
150
0
1.5
50
120

Runways
slight/moderate
extensive
complete
2.5
35
85
2.5
35
65

784
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
Table 7.16.b Discretized Restoration Functions for Aiport Sub-Components
Discretized Restoration Functions
Classification Damage State 1 day 3 days 7 days 30 days 90 days
Control Towers,
Parking Structures,
Hangar Facilities,
Terminal Building
slight
moderate
extensive
complete
100
37
16
11
100
84
17
11
100
100
20
12
100
100
34
16
100
100
79
31

Runways
slight/moderate
extensive
complete
27
17
10
57
18
11
100
21
12
100
44
20
100
95
53

7.7.8 Development of Damage Functions

Damage functions for airport system facilities are defined in terms of PGA and PGD
except for runways (PGD only). Note that, unless it is specified otherwise, ground
failure (PGD) related damage functions for these facilities are assumed to be similar to
those described for railroad system facilities in section 7.2.8.

An example of how to combine PGD and PGA algorithms is presented in section 7.2.8.

Damage Functions for Runways

The earthquake hazard for airport runways is ground failure. Little damage is
attributed to ground shaking; therefore, the damage function includes only ground
failure as the hazard. All runways are assumed to be paved. The median values and
dispersion for the damage states for runways are given in Table 7.17. These damage
functions are also shown in Figure 7.32.

Table 7.17 Damage Algorithms for Runways
Permanent Ground Deformation
Classification Damage State Median (in) |

Runways
slight/moderate
extensive
complete
1
4
12
0.6
0.6
0.6

Damage Functions for Rest of Airport System Components

In section 7.7.5, these components were defined by "one to one" correspondence with
those for railway systems. Therefore, damage functions for the remaining airport
components (i.e. fuel facilities, maintenance facilities, and other buildings) can be
found in Section 7.2.8.




785
HazusMHTechnicalManual
7.7.9 Guidance for Loss Estimation Using Advanced Data and Models Analysis

For this level of analysis, the expert can use the methodology developed with the
flexibility to: (1) include a refined inventory of the airport system pertaining to the area
of study, and (2) include component specific and system specific fragility data. Default
User-Supplied Data Analysis damage algorithms can be modified or replaced to
accommodate any specified key component of a airport system, such as a control tower.
Similarly, better restoration curves could be developed given knowledge of available
resources and a more accurate layout of the transportation network within the local
topographic and geological conditions.

7.7.10 References

Applied Technology Council, "Earthquake Damage Evaluation Data for California",
ATC-13, Redwood City, CA, 1985.

G & E Engineering Systems, Inc. (G & E), "NIBS Earthquake Loss Estimation Methods,
Technical Manual, Transportation Systems (Airport Systems)", May 1994.


786
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems

Time (days)

P
e
r
c
e
n
t

F
u
n
c
t
i
o
n
a
l

0
25
50
75
100
1 10 100 1000
Minor Moderate Extensive Complete ATC-13 Data

Figure 7.30 Restoration Curve for Airport Runways.


Time (days)

P
e
r
c
e
n
t

F
u
n
c
t
i
o
n
a
l

0
25
50
75
100
1 10 100 1000
Minor Moderate Extensive Complete ATC-13 Data

Figure 7.31 Restoration Curves for Airport Buildings, Facilities, and Control
Tower.




787
HazusMHTechnicalManual






Permanent Ground Deformation (inches)
P
r
o
b
a
b
i
l
i
t
y

[

D
s


>


d
s


|


P
G
D

]






0
0.25
0.5
0.75
1
0.00 6.00 12.00 18.00 24.00 30.00
Slight/Moderate Extensive Complete


Figure 7.32 Fragility Curves for Runways Subject to Permanent Ground
Deformation at Various Damage States.



Hazus-MH Technical Manual
APPENDIX 7A


Any given subcomponent in the lifeline methodology can experience all five damage
states; however, the only damage states listed in the appendices of Chapters 7 and 8 are
the ones used in the fault tree logic of the damage state of interest of the component.

Table A.7.1 Subcomponent Damage Algorithms: Rock Tunnels
(after G&E, 1994)
Peak Ground Acceleration
Subcomponents Damage State Median (g) |
Liner
slight
moderate
0.6
0.8
0.4
0.6

Permanent Ground Deformation
Subcomponents Damage State Median (in) |
Liner
slight
extensive
complete
6
12
60
0.7
0.5
0.5
Portal
slight
extensive
complete
6
12
60
0.7
0.5
0.5


Table A.7.2 Subcomponent Damage Algorithms: Cut & Cover Tunnels
(after G&E, 1994)
Peak Ground Acceleration
Subcomponents Damage State Median (g) |
Liner
slight
moderate
0.5
0.7
0.4
0.6

Permanent Ground Deformation
Subcomponents Damage State Median (in) |
Liner
slight
extensive
complete
6
12
60
0.7
0.5
0.5
Portal
slight
extensive
complete
6
12
60
0.7
0.5
0.5


790
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
APPENDIX 7B




Table B.7.1 Subcomponent Damage Algorithms:
Seismically Designed Railway Bridges (after G&E, 1994)
Peak Ground Acceleration
Subcomponents Damage State Median (g) |
Column
slight
extensive
complete
0.45
1.0
1.4
0.55
0.7
0.7
Abutment
slight
moderate
0.45
1.0
0.55
0.7
Connection
moderate
extensive
0.86
1.4
0.70
0.70
Deck slight 0.67 0.55

Permanent Ground Deformation
Subcomponents Damage State Median (in) |
Column
extensive
complete
14
28
0.7
0.7
Abutment
moderate
extensive
15
30
0.7
0.7
Connection complete 30 0.7
Approach
slight
moderate
extensive
2
12
24
0.5
0.7
0.7






791
Hazus-MH Technical Manual

Table B.7.2 Subcomponent Damage Algorithms:
Conventionally Designed Railway Bridges (after G&E, 1994)
Peak Ground Acceleration
Subcomponents Damage State Median (g) |
Column
slight
extensive
complete
0.3
0.8
1.0
0.55
0.7
0.7
Abutment
slight
moderate
0.3
0.8
0.55
0.7
Connection
moderate
extensive
0.7
1.0
0.70
0.70
Deck slight 0.5 0.55

Permanent Ground Deformation
Subcomponents Damage State Median (in) |
Column
extensive
complete
10
21
0.7
0.7
Abutment
moderate
extensive
10
21
0.7
0.7
Connection complete 21 0.7
Approach
slight
moderate
extensive
2
12
24
0.5
0.7
0.7






792
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
Table B.7.3 Subcomponent Damage Algorithms:
Fuel Facility with Anchored Components (after G&E, 1994)
Peak Ground Acceleration
Subcomponents Damage State Median (g)
|
Electric Power
(Backup)
slight
moderate
0.80
1.00
0.60
0.80
Electric Power
(Off-Site)
slight
moderate
0.15
0.25
0.6
0.5
Tank
slight
moderate
extensive
complete
0.30
0.70
1.25
1.60
0.60
0.60
0.65
0.60
Pump Building
slight
moderate
extensive
complete
0.15
0.40
0.80
1.50
0.80
0.80
0.80
0.80
Horizontal
Pump
extensive 1.60 0.60
Equipment moderate 1.00 0.60



Table B.7.4 Subcomponent Damage Algorithms:
Fuel Facility with Unanchored Components (after G&E, 1994)
Peak Ground Acceleration
Subcomponents Damage State Median (g)
|
Electric Power
(Backup)
slight
moderate
0.20
0.40
0.60
0.80
Electric Power
(Off-Site)
slight
moderate
0.15
0.25
0.6
0.5
Tank
slight
moderate
extensive
complete
0.15
0.35
0.68
0.95
0.70
0.75
0.75
0.70
Pump Building
slight
moderate
extensive
complete
0.15
0.40
0.80
1.50
0.80
0.80
0.80
0.80
Horizontal
Pump
extensive 1.60 0.60
Equipment moderate 0.60 0.60




793
Hazus-MH Technical Manual
Table B.7.5 Subcomponent Damage Algorithms:
Dispatch Facility with Anchored Components (after G&E, 1994)
Peak Ground Acceleration
Subcomponents Damage State Median (g)
|
Electric Power
(Backup)
slight
moderate
0.80
1.00
0.60
0.80
Electric Power
(Off-Site)
slight
moderate
0.15
0.25
0.6
0.5
Building
slight
moderate
extensive
complete
0.15
0.40
0.80
1.50
0.80
0.80
0.80
0.80
Equipment moderate 1.00 0.60




Table B.7.6 Subcomponent Damage Algorithms:
Dispatch Facility with Unanchored Components (after G&E, 1994)
Peak Ground Acceleration
Subcomponents Damage State Median (g)
|
Electric Power
(Backup)
slight
moderate
0.20
0.40
0.60
0.80
Electric Power
(Off-Site)
slight
moderate
0.15
0.25
0.6
0.5
Building
slight
moderate
extensive
complete
0.15
0.40
0.80
1.50
0.80
0.80
0.80
0.80
Equipment moderate 0.60 0.60





794
Chapter7DirectPhysicalDamagetoLifelinesTransportationSystems
APPENDIX 7C




Table C.7.1 Subcomponent Damage Algorithms for DC Power
Substation with Anchored Components
Peak Ground Acceleration
Subcomponents Damage State Median (g) |
Building
slight
moderate
extensive
complete
0.15
0.40
0.80
1.50
0.80
0.80
0.80
0.80
Equipment moderate 1.00 0.60
Off-Site Power
slight
moderate
0.15
0.25
0.6
0.5






Table C.7.2 Subcomponent Damage Algorithms for DC Power
Substation with Unanchored Components
Peak Ground Acceleration
Subcomponents Damage State Median (g) |
Building
slight
moderate
extensive
complete
0.15
0.40
0.80
1.50
0.80
0.80
0.80
0.80
Equipment moderate 0.60 0.60
Off-Site Power
slight
moderate
0.15
0.25
0.6
0.5




795
Hazus-MH Technical Manual
APPENDIX 7D





Table 7.D.1 Subcomponent Damage Algorithms for Waterfront Structures
Permanent Ground Deformation
Subcomponents Damage State Median (in) |
Wharf slight 8 0.6
Piers
slight
moderate
extensive
complete
8
16
24
60
0.6
0.6
0.6
0.6
Seawalls
slight
moderate
extensive
complete
8
16
24
60
0.6
0.6
0.6
0.6



81
HazusMHTechnicalManual
Chapter 8
Direct Damage to Lifelines - Utility Systems

This chapter describes and presents the methodology for estimating direct damage to
Utility Systems. The Utility Module is composed of the following six systems:

Potable Water
Waste Water
Oil (crude and refined)
Natural Gas
Electric Power
Communication

The flowchart of the overall methodology, highlighting the utility system module and its
relationship to other modules, is shown in Flowchart 8.1.

8.1 Potable Water Systems

8.1.1 Introduction

This section presents a loss estimation methodology for a water system during
earthquakes. This system consists of supply, storage, transmission, and distribution
components. All of these components are vulnerable to damage during earthquakes,
which may result in a significant disrutpion to the water utility network.

8.1.2 Scope

The scope of this section includes development of methods for estimation of earthquake
damage to a potable water system given knowledge of the system's components (i.e.,
tanks, aqueducts, water treatment plants, wells, pumping stations, conveyance pipes,
junctions, hydrants, and valves), classification (i.e., for water treatment plants, small,
medium or large), and the ground motion (i.e. peak ground velocity, peak ground
acceleration and/or permanent ground deformation). Damage states describing the level
of damage to each of the water system components are defined (i.e., slight/minor,
moderate, extensive, or complete), while for pipelines, the number of repairs/km is the
key paramter. Fragility curves are developed for each classification of the water system
component. These curves describe the probability of reaching or exceeding each damage
state given the level of ground motion or ground failure.

Based on these fragility curves, a method for assessing functionality of each component
of the water system is presented. A simplified approach for evaluating the overall water
system network performance is also provided.




82
Chapter8DirectDamagetoLifelinesUtilitySystems

8. Lifelines-
Utility
Systems
4. Ground Motion 4. Ground Failure
Direct Physical
Damage
6. Essential and
High Potential
Loss Facilities
12. Debris 10. Fire 15. Economic 14. Shelter 9. Inundation 11. HazMat
16. Indirect
Economic
Losses
Potential Earth Science Hazards
Direct Economic/
Social Losses
Induced Physical
Damage
7. Lifelines-
Transportation
Systems
5. General
Building
Stock
13. Casualities



Flowchart 8.1 Utility System Damage Relationship to Other Modules of the
Earthquake Loss Estimation Methodology



83
HazusMHTechnicalManual
8.1.3 Input Requirements and Output Information

Depending on the desired level of analysis, the input required for analyzing water
systems varies. In total, three levels of analysis are enabled in Hazus.

Level One:
The default inventory in Hazus contains estimate of potable water pipelines aggregated at
the census tract level. This pipeline data was developed using the US Census TIGER
street file datasets. For the level one analysis, eighty (80) percent of the pipes are
assumed to be brittle with the remaining pipes assumed to be ductile. In addition, peak
ground velocity and permanent ground deformation (PGV and PGD) for each census tract
is needed for the analysis.

The results from a level one analysis include the expected number of leaks and breaks per
census tract and a simplified evaluation of the potable water system network performance
(i.e. number of households without water).

Level Two:
For this level, the input required to estimate damage to potable water systems includes
the following items:

Transmission Aqueducts and Distribution Pipelines

Geographical location of aqueduct/pipe links (longitude and latitude of end nodes)
Peak ground velocity and permanent ground deformation (PGV and PGD)
Classification (ductile pipe or brittle pipe)

Reservoirs, Water Treatment Plants, Wells, Pumping Stations and Storage Tanks

Geographical location of facility (longitude and latitude)
PGA and PGD
Classification (e.g., capacity and anchorage)

Direct damage output from level 2 analysis includes probability estimates of (1)
component functionality and (2) damage, expressed in terms of the component's damage
ratio (repair cost to replacement cost). Note that damage ratios for each of the potable
water system components are presented in section 15.3 of Chapter 15. In addition, a
simplified evaluation of the potable water system network performance is also provided.
This is based on network analyses done for Oakland, San Francisco and Tokyo. The
output from this simplified version of network analysis consists of an estimate of the flow
reduction to the areas served by the water system being evaluated. Details of this
methodology are presented in subsection 8.1.9.

84
Chapter8DirectDamagetoLifelinesUtilitySystems
Level Two Enhanced:
This level of analysis essentially relies on the same type of information provided in the
previous level with four main differences:

Three additional components are considered. These are: junctions, hydrants, and
valves.
Connectivity of the components is maintained (i.e., what facilities are connected
to which pipeline links or valves).
Serviceability in the system considered (i.e., the demand pressures and flow
demands at the different distribution nodes).
Input data for the water system need to be in one of the following three
commercially available formats: KYPIPE, EPANET, or CYBERNET.

Recent work by Khater and Waisman (EQE, 1999) elaborates in great details on the level
two enhanced analysis model implemented in Hazus

. In particular, this work provides a


comprehensive theoretical background on the governing equations for a water system and
explains how the commercial data need to be formatted in order to be able to import it
into Hazus

. This work is available in a separate document entitled Potable Water


System Analysis Model (POWSAM) that can be acquired directly from NIBS.

Results from the level two enhanced analysis are similar to the level two. That is,
probability estimates of (1) component functionality and (2) damage, expressed in terms
of the component's damage ratio (repair cost to replacement cost). The main difference is
in the evaluation of the potable water system network performance, which is in this case
based on a more comprehensive approach. Note that in either case, the performance is
expressed in terms of an estimate of the flow reduction to the areas served by the water
system being evaluated and the number of households expected to be deprived from
water.


8.1.4 Form of Damage Functions

Damage functions or fragility curves for water system components other than pipelines
are modeled as lognormally-distributed functions that give the probability of reaching or
exceeding different damage states for a given level of ground motion (quantified in terms
of PGA) and ground failure (quantified in terms of PGD). Each of these fragility curves
is characterized by a median value of ground motion (or failure) and an associated
dispersion factor (lognormal standard deviation). For pipelines, empirical relations that
give the expected repair rates due to ground motion (quantified in terms of PGV) or
ground failure (quantified in terms of PGD) are provided. Definitions of various damage
states and the methodology used in deriving all these fragility curves are presented in the
next section.

85
HazusMHTechnicalManual
8.1.5 Description of Potable Water System Components

A potable water system typically consists of terminal reservoirs, water treatment plants,
wells, pumping plants, storage tanks and transmision and distribution pipelines. In this
subsection, a brief description of each of these components is presented.

Terminal Reservoirs

Terminal reservoirs are typically lakes (man made or natural) and are usually located
nearby and upstream of the water treatment plant. Vulnerability of terminal reservoirs
and associated dams is marginally assessed in the loss estimation methodology.
Therefore, even though reservoirs are an essential part of a potable water system, it is
assumed in the analysis of water systems that the amount of water flowing into water
treatment plants from reservoirs right after an earthquake is essentially the same as before
the earthquake.

Transmission Aqueducts

These transmission conduits are typically large size pipes (more than 20 inches in
diameter) or channels (canals) that convey water from its source (reservoirs, lakes, rivers)
to the treatment plant.

Transmission pipelines are commonly made of concrete, ductile iron, cast iron, or steel.
These could be elevated/at grade or buried. Elevated or at grade pipes are typically made
of steel (welded or riveted), and they can run in single or multiple lines.

Canals are typically lined with concrete, mainly to avoid excessive loss of water by
seepage and to control erosion. In addition to concrete lining, expansion joints are
usually used to account for swelling and shrinkage under varying temperature and
moisture conditions. Damageability of channels has occurred in some earthquake, but is
outside the scope of the scope of the methodology.

Supply Facilities- Water Treatment Plants (WTP)

Water treatment plants are generally composed of a number of physical and chemical unit
processes connected in series, for the purpose of improving the water quality. A
conventional WTP consists of a coagulation process, followed by a sedimentation
process, and finally a filtration process. Alternately, a WTP can be regarded as a system
of interconnected pipes, basins, and channels through which the water moves, and where
the flow is governed by hydraulic principles. WTP are categorized as follows:

Small water treatment plants, with capacity ranging from 10 mgd to 50 mgd, are assumed
to consist of a filter gallery with flocculation tanks (composed of paddles and baffles) and
settling (or sedimentation) basins as main components, chemical tanks (needed in the
coagulation and other destabilization processes), chlorination tanks, electrical and
mechanical equipment, and elevated pipes.
86
Chapter8DirectDamagetoLifelinesUtilitySystems

Medium water treatment plants, with capacity ranging from 50 mgd to 200 mgd, are
simulated by adding more redundancy to small treatment plants (i.e. twice as many
flocculation, sedimentation, chemical and chlorination tanks).

Large water treatment plants, with capacity above 200 mgd, are simulated by adding even
more redundancy to small treatment plants (i.e., three times as many flocculation,
sedimentation, chemical and chlorination tanks/basins).

Water treatment plants are also classified based on whether the subcomponents
(equipment and backup power) are anchored or not as defined in section 7.2.5.

Pumping Plants (PP)

Pumping plants are usually composed of a building, one or more pumps, electrical
equipment, and in some cases, backup power systems. Pumping plants are classified as
either small PP (less than 10 mgd capacity) or medium/large PP (more than 10 mgd
capacity). Pumping plants are also classified with respect to whether the subcomponents
(equipment and backup power) are anchored or not. As noted in Chapter 7, anchored
means equipment designed with special seismic tie downs and tiebacks while unanchored
means equipment with manufactures normal requirements.

Wells (WE)

Wells typically have a capacity between 1 and 5 mgd. Wells are used in many cities as
a primary or supplementary source of water supply. Wells include a shaft from the
surface down to the aquifer, a pump to bring the water up to the surface, equipment used
to treat the water, and sometimes a building, which encloses the well and equipment.

Water Storage Tanks (ST)

Water storage tanks can be elevated steel, on ground steel (anchored/unanchored), on
ground concrete (anchored/unanchored), buried concrete, or on ground wood tanks.
Typical capacity of storage tanks is in the range of 0.5 mgd to 2 mgd.

Distribution Facilities and Distribution Pipes

Distribution of water can be accomplished by gravity, or by pumps in conjunction with
on-line storage. Except for storage reservoirs located at a much higher altitude than the
area being served, distribution of water would necessitate, at least, some pumping along
the way. Typically, water is pumped at a relatively constant rate, with flow in excess of
consumption being stored in elevated storage tanks. The stored water provides a reserve
for fire flow and may be used for general-purpose flow should the electric power fail, or
in case of pumping capacity loss.

87
HazusMHTechnicalManual
Distribution pipelines are commonly made of concrete (prestressed or reinforced),
asbestos cement, ductile iron, cast iron, steel, or plastic. The selection of material type
and pipe size are based on the desired carrying capacity, availability of material,
durability, and cost. Distribution pipes represent the network that delivers water to
consumption areas. Distribution pipes may be further subdivided into primary lines,
secondary lines, and small distribution mains. The primary or arterial mains carry flow
from the pumping station to and from elevated storage tanks, and to the consumption
areas, whether residential, industrial, commercial, or public. These lines are typically
laid out in interlocking loops, and all smaller lines connecting to them are typically
valved so that failure in smaller lines does not require shutting off the larger. Primary
lines can be up to 36 inches in diameter. Secondary lines are smaller loops within the
primary mains and run from one primary line to another. They serve primarily to provide
a large amount of water for fire fighting without excessive pressure loss. Small
distribution lines represent the mains that supply water to the user and to the fire
hydrants.

In this earthquake loss estimation study, the simplified method for water system network
performance evaluation applies to a distribution pipe network digitized at the primary
level.

8.1.6 Definition of Damage States

Potable water systems are susceptible to earthquake damage. Facilities such as water
treatment plants; wells, pumping plants and storage tanks are most vulnerable to PGA,
and sometimes PGD, if located in liquefiable or landslide zones. Therefore, the damage
states for these components are defined and associated with PGA and PGD. Aqueducts
and pipelines, on the other hand, are vulnerable to PGV and PGD. Therefore, the damage
states for these components are associated with these two ground motion parameters.

8.1.6.1 Damage State Defintions for Components Other than Pipelines

A total of five damage states for potable water system components are defined. These are
none (ds
1
), slight/minor (ds
2
), moderate (ds
3
), extensive (ds
4
), and complete (ds
5
).
88
Chapter8DirectDamagetoLifelinesUtilitySystems
Slight/Minor Damage (ds
2
)

For water treatment plants, ds
2
is defined by malfunction of plant for a short time
(less than three days) due to loss of electric power and backup power if any,
considerable damage to various equipment, light damage to sedimentation basins,
light damage to chlorination tanks, or light damage to chemical tanks. Loss of water
quality may occur.

For pumping plants, ds
2
is defined by malfunction of plant for a short time (less than
three days) due to loss of electric power and backup power if any, or slight damage to
buidings.

For wells, ds
2
is defined by malfunction of well pump and motor for a short time
(less than three days) due to loss of electric power and backup power if any, or light
damage to buidings.

For Storage Tanks, ds
2
is defined by the tank suffering minor damage without loss
of its contents or functionality. Minor damage to the tank roof due to water sloshing,
minor cracks in concrete tanks, or localized wrinkles in steel tanks fits the description
of this damage state.

Moderate Damage (ds
3
)

For water treatment plants, ds
3
is defined by malfunction of plant for about a week
due to loss of electric power and backup power if any, extensive damage to various
equipment, considerable damage to sedimentation basins, considerable damage to
chlorination tanks with no loss of contents, or considerable damage to chemical tanks.
Loss of water quality is imminent.

For pumping plants, ds
3
is defined by the loss of electric power for about a week,
considerable damage to mechanical and electrical equipment, or moderate damage to
buildings.

For wells, ds
3
is defined by malfunction of well pump and motor for about a week
due to loss of electric power and backup power if any, considerable damage to
mechanical and electrical equipment, or moderate damage to buildings.

For Storage Tanks, ds
3
is defined by the tank being considerably damaged, but only
minor loss of content. Elephant foot buckling for steel tanks without loss of content,
or moderate cracking of concrete tanks with minor loss of content fits the description
of this damage state.

89
HazusMHTechnicalManual
Extensive Damage (ds
4
)

For water treatment plants, ds
4
is defined by the pipes connecting the different
basins and chemical units being extensively damaged. This type of damage will
likely result in the shutdown of the plant.

For pumping plants, ds
4
is defined by the building being extensively damaged, or
the pumps being badly damaged beyond repair.

For wells, ds
4
is defined by the building being extensively damaged or the well pump
and vertical shaft being badly distorted and nonfunctional.

For Storage Tanks, ds
4
is defined by the tank being severely damaged and going out
of service. Elephant foot buckling for steel tanks with loss of content, stretching of
bars for wood tanks, or shearing of wall for concrete tanks fits the description of this
damage state.

Complete Damage (ds
5
)

For water treatment plants, ds
5
is defined by the complete failure of all pipings, or
extensive damage to the filter gallery.

For pumping plants, ds
5
is defined by the building collapsing.

For wells, ds
5
is defined by the building collapsing.

For Storage Tanks, ds
5
is defined by the tank collapsing and losing all of its content.

8.1.6.2 Defintion of Damage States for Pipelines

For pipelines, two damage states are considered. These are leaks and breaks. Generally,
when a pipe is damaged due to ground failure (PGD), the type of damage is likely to be a
break, while when a pipe is damaged due to seismic wave propagation (PGV), the type of
damage is likely to be joint pull-out or crushing at the bell. In the loss methodology, it is
assumed that damage due to seismic waves will consist of 80% leaks and 20% breaks,
while damage due to ground failure will consist of 20% leaks and 80% breaks. The user
can override these default percentages.

8.1.7 Component Restoration Curves

Restoration functions for potable water system components, namely, water treatment
plants, wells, pumping plants, and storage tanks are based on SF-30a, SF-30b and SF-30d
of ATC-13 consistent with damage states defined in the previous section. That is,
restoration functions for ds
2
, ds
3
, ds
4
, and ds
5
defined herein are assumed to correspond
810
Chapter8DirectDamagetoLifelinesUtilitySystems
to ds
2
, ds
3
, ds
4
, and ds
5
of ATC-13. The parameters of these restoration curves are given
in Tables 8.1.a and 8.1.b, and 8.1.c.

Table 8.1.a: Continuous Restoration Functions for Potable Water Systems
(After ATC-13, 1985)
Restoration Functions (All Normal Distributions)
Classification Damage State Mean (Days) (days)
Water Treatment
Plants
slight/minor
moderate
extensive
complete
0.9
1.9
32.0
95.0
0.3
1.2
31.0
65.0

Pumping Plants
slight/minor
moderate
extensive
complete
0.9
3.1
13.5
35.0
0.3
2.7
10.0
18.0

Wells
slight/minor
moderate
extensive
complete
0.8
1.5
10.5
26.0
0.2
1.2
7.5
14.0
Water Storage
Tanks
slight/minor
moderate
extensive
complete
1.2
3.1
93.0
155.0
0.4
2.7
85.0
120.0

Table 8.1.a gives means and standard deviations for each restoration curve (i.e., smooth
continuous curve), while Table 8.1.b gives approximate discrete functions for the
restoration curves developed. These restoration functions are also shown in Figures 8.1
through 8.4.


Table 8.1.b: Discretized Restoration Functions for Potable Water System
Components
Discretized Restoration Functions
Classification Damage State 1 day 3 days 7 days 30 days 90 days
Water Treatment
Plants
slight/minor
moderate
extensive
complete
65
23
16
7
100
82
18
8
100
100
21
9
100
100
48
16
100
100
97
47

Pumping Plants
slight/minor
moderate
extensive
complete
65
22
10
3
100
50
15
4
100
93
25
6
100
100
95
40
100
100
100
100
811
HazusMHTechnicalManual
Table 8.1.b: Discretized Restoration Functions for Potable Water System
Components (continued)
Discretized Restoration Functions
Classification Damage State 1 day 3 days 7 days 30 days 90 days

Wells
slight/minor
moderate
extensive
complete
85
34
11
4
100
90
16
6
100
100
33
9
100
100
100
62
100
100
100
100
Water Storage
Tanks
slight/minor
moderate
extensive
complete
30
20
13
10
100
49
15
11
100
93
16
12
100
100
23
15
100
100
49
30

The restoration functions for pipelines are expressed in terms of number of days needed
to fix the leaks and breaks. These restoration functions are given in Table 8.1.c

Table 8.1.c: Restoration Functions for Potable Water Pipelines
Class

Diameter
from: [in]

Diameter to:
[in]
# Fixed
Breaks per
Day per
Worker
# Fixed
Leaks per
Day per
Worker

# Available
Workers
Priority
a 60 300 0.33 0.66 User-
specified
1 (Highest)
b 36 60 0.33 0.66 User-
specified
2
c 20 36 0.33 0.66 User-
specified
3
d 12 20 0.50 1.0 User-
specified
4
e 8 12 0.50 1.0 User-
specified
5 (Lowest)
u
Unknown
diameter
or for Default
Data Analysis
0.50 1.0 User-
specified
6 (lowest)

Where the total number of available workers can be specified by the user. It should be
noted that the values in Table 8.1.c are based on the following 4 assumptions:

(1) Pipes that are less than 20 in diameter are defined as small, while pipes with
diameter greater than 20 are defined as large.
(2) For both small and large pipes a 16 hour day shift is assumed.
(3) For small pipes, a 4-person crew needs 4 hours to fix a leak, while the same 4-person
crew needs 8 hours to fix a break. (Mathematically, this is equivalent to saying it takes 16
people to fix a leak in one hour and it takes 32 people to fix a break in one hour).
(4) For large pipes, a 4-person crew needs 6 hours to fix a leak, while the same 4-person
crew needs 12 hours to fix a break. (Mathematically, this is equivalent to say it takes 24
people to fix a leak in one hour and 48 people to fix a break in one hour).
812
Chapter8DirectDamagetoLifelinesUtilitySystems
With this algorithm for potable water pipelines, the total number of days needed to finish
repairs is calculated as:

Days needed to finish all repairs =(1/available work) * [(#small pipe leaks/1.0) +(#
small pipe breaks/0.5) +(#large pipe leaks/0.66) +(#large pipe breaks/0.33)]

The percentage of repairs finished at Day1, Day3, Day7, Day30, and Day90 are then
computed using linear interpolation.

8.1.8 Development of Damage Functions

In this subsection, damage functions for the various components of a potable water
system are presented. In cases where the components are made of subcomponents (i.e.,
water treatment plants, pumping plants, and wells), fragility curves for these components
are based on the probabilistic combination of subcomponent damage functions using
Boolean expressions to describe the relationship of subcomponents to the components. It
should be mentioned that the Boolean logic is implicitly presented within the definition
of a particular damage state. For example, slight/minor damage for a water treatment
plant was defined by malfunction for a short time due to loss of electric power AND
backup power (if any), considerable damage to various equipment, light damage to
sedimentation basins, light damage to chlorination tanks, OR light damage to chemical
tanks. Therefore, the fault tree for slight/minor damage has FIVE primary OR branches:
electric power, equipment, sedimentation basins, chlorination tanks, and chemical tanks,
and TWO secondary AND branches under electric power: commercial power and backup
power. The Boolean approach involves evaluation of the probability of each component
reaching or exceeding different damage states, as defined by the damage level of its
subcomponents. These evaluations produce component probabilities at various levels of
ground motion. In general, the Boolean combinations do not produce a lognormal
distribution, so a lognormal curve that best fits this probability distribution is determined
numerically. It should be mentioned that damage functions due to ground failure (i.e.,
PGD) for all potable water systems components except pipelines (i.e., water treatment
plants, pumping plants, wells, and storage tanks) are assumed to be similar to those
described for buildings, unless specified otherwise. These are:

- For lateral spreading, a lognormal damage function with a median of 60 inches and a
dispersion of 1.2 is assumed for the damage state of "at least extensive". 20% of this
damage is assumed to be complete. For a PGD of 10 inches due to lateral spreading,
there is a 7% probability of "at least extensive" damage.

For vertical settlement, a lognormal curve with a median of 10 inches and a dispersion of
1.2 is assumed for the damage state of "at least extensive". 20% of this damage is
assumed to be complete. For a PGD of 10" due to vertical settlement, there is a 50%
chance of "at least extensive" damage.

813
HazusMHTechnicalManual
- For fault movement or landslide, a lognormal curve with a median of 10 inches and a
dispersion of 0.5 is assumed for "complete" damage state. That is, for 10 inches of PGD
due to fault movement or landslide, there is a 50% chance of "complete" damage.

An example of how to combine a PGD algorithm with a PGA algorithm for lifeline
components was presented in section 7.2.8 of Chapter 7.

Damage Functions for Water Treatment Plants (due to Ground Shaking)

PGA related damage functions for water treatment plants are developed with respect to
their classification. A total of 24 damage functions are presented. Half of these damage
functions correspond to water treatment plants with anchored subcomponents, while the
other half correspond to water treatment plants with unanchored subcomponents (see
section 7.2.5 for the definition of anchored and unanchored subcomponents). Medians
and dispersions of these damage functions are given in Tables 8.3 through 8.5.

Medians and dispersions of damage functions for the water treatment plant
subcomponents are summarized in Tables A.8.6 and A.8.7 of Appendix 8A. The
medians for elevated pipe damage functions in these tables are based on ATC-13 data
(FC-32) for "at grade pipe" using the following MMI to PGA conversion (after G&E,
1994), along with a best-fit lognormal curve.

Table 8.2: MMI to PGA Conversion (after G&E, 1994)
MMI VI VII VIII IX X XI XII
PGA
0.12 0.21 0.36 0.53 0.71 0.86 1.15

Graphical representations of water treatment plant damage functions are also provided.
Figures 8.5 through 8.10 are fragility curves for the different classes of water treatment
plants.

814
Chapter8DirectDamagetoLifelinesUtilitySystems
Table 8.3: Damage Algorithms for Small Water Treatment Plants
Peak Ground Acceleration
Classification Damage State Median (g)
Plants with
anchored
subcomponents
( PWT1)
slight/minor
moderate
extensive
complete
0.25
0.38
0.53
0.83
0.50
0.50
0.60
0.60
Plants with
unanchored
subcomponents
(PWT2)
slight/minor
moderate
extensive
complete
0.16
0.27
0.53
0.83
0.40
0.40
0.60
0.60

Table 8.4: Damage Algorithms for Medium Water Treatment Plants
Peak Ground Acceleration
Classification Damage State Median (g)
Plants with
anchored
subcomponents
(PWT3)
slight/minor
moderate
extensive
complete
0.37
0.52
0.73
1.28
0.40
0.40
0.50
0.50
Plants with
unanchored
subcomponents
(PWT4)
slight/minor
moderate
extensive
complete
0.20
0.35
0.75
1.28
0.40
0.40
0.50
0.50

Table 8.5: Damage Algorithms for Large Water Treatment Plants
Peak Ground Acceleration
Classification Damage State Median (g)
Plants with
anchored
subcomponents
(PWT5)
slight/minor
moderate
extensive
complete
0.44
0.58
0.87
1.57
0.40
0.40
0.45
0.45
Plants with
unanchored
subcomponents
(PWT6)
slight/minor
moderate
extensive
complete
0.22
0.35
0.87
1.57
0.40
0.40
0.45
0.45

Damage Functions for Pumping Plants (due to Ground Shaking)

PGA related damage functions for pumping plants are developed with respect to their
classification. A total of 16 damage functions are presented. Half of these damage
functions correspond to pumping plants with anchored subcomponents, while the
other half correspond to pumping plants with unanchored subcomponents. Medians
and dispersions of these damage functions are given in Tables 8.6 and 8.7. Graphical
representations of damage functions for the different classes of pumping plants are
presented in Figures 8.11 through 8.14. Note that medians and dispersions of damage
functions for pumping plants' subcomponents are summarized in Appendix 8A.
815
HazusMHTechnicalManual

Table 8.6: Damage Algorithms for Small Pumping Plants
Peak Ground Acceleration
Classification Damage State Median (g)

Plants with
anchored
subcomponents
(PPP1)
slight/minor
moderate
extensive
complete
0.15
0.36
0.66
1.50
0.70
0.65
0.65
0.80
Plants with
unanchored
subcomponents
(PPP2)
slight/minor
moderate
extensive
complete
0.13
0.28
0.66
1.50
0.60
0.50
0.65
0.80

Table 8.7: Damage Algorithms for Medium/Large Pumping Plants
Peak Ground Acceleration
Classification Damage State Median (g)
Plants with
anchored
subcomponents
(PPP3)
slight/minor
moderate
extensive
complete
0.15
0.36
0.77
1.50
0.75
0.65
0.65
0.80
Plants with
unanchored
subcomponents
(PPP4)
slight/minor
moderate
extensive
complete
0.13
0.28
0.77
1.50
0.60
0.50
0.65
0.80

Damage Functions for Wells (due to Ground Shaking)

A total of four PGA-related damage functions are presented. In developing these
damage functions, it is assumed that equipment in wells is anchored. Medians and
dispersions of these damage functions are given in Table 8.8. Graphical
representations of well damage functions are also shown in Figure 8.15. Note that
medians and dispersions of damage functions for well subcomponents are
summarized in Appendix 8A.

Table 8.8: Damage Algorithms for Wells
Peak Ground Acceleration
Classification Damage State Median (g)

Wells (PWE1)
slight/minor
moderate
extensive
complete
0.15
0.36
0.72
1.50
0.75
0.65
0.65
0.80

816
Chapter8DirectDamagetoLifelinesUtilitySystems
Damage Functions for Water Storage tanks

A total of 24 PGA related damage functions are developed. These correspond to on-
ground concrete (anchored and unanchored), on ground steel (anchored and unanchored),
elevated steel, and on-ground wood tanks. For tanks, anchored and unanchored refers to
positive connection, or a lack thereof, between the tank wall and the supporting concrete
ring wall. The PGD algorithm associated with these water storage tanks is described at
the beginning of section 8.1.8. For buried storage tanks a separate PGD algorithm is
presented. Medians and dispersions of the PGA related damage functions are given in
Table 8.9. Graphical representations of water storage tank damage functions are also
provided. Figures 8.16 through 8.21 are fragility curves for the different classes of water
storage tanks.

Table 8.9: Damage Algorithms for Water Storage Tanks
Peak Ground Acceleration
Classification Damage State Median (g)
On-Ground
Anchored
Concrete Tank
(PST1)
slight/minor
moderate
extensive
complete
0.25
0.52
0.95
1.64
0.55
0.70
0.60
0.70
On-Ground
Unanchored
Concrete Tank
(PST2)
slight/minor
moderate
extensive
complete
0.18
0.42
0.70
1.04
0.60
0.70
0.55
0.60
On-Ground
Anchored Steel
Tank
(PST3)
slight/minor
moderate
extensive
complete
0.30
0.70
1.25
1.60
0.60
0.60
0.65
0.60
On-Ground
Unanchored
Steel Tank
(PST4)
slight/minor
moderate
extensive
complete
0.15
0.35
0.68
0.95
0.70
0.75
0.75
0.70
Above-Ground
Steel Tank
(PST5)
slight/minor
moderate
extensive
complete
0.18
0.55
1.15
1.50
0.50
0.50
0.60
0.60
On-Ground
Wood Tank
(PST6)
slight/minor
moderate
extensive
complete
0.15
0.40
0.70
0.90
0.60
0.60
0.70
0.70
Permanent Ground Deformation
Classification Damage State Median (in)

Buried Concrete
Tank (PST7)
slight/minor
moderate
extensive
complete
2
4
8
12
0.50
0.50
0.50
0.50

817
HazusMHTechnicalManual
Damage Functions for Buried Pipelines

Two damage algorithms are used for buried pipelines. The first algorithm is
associated with peak ground velocity (PGV) while the second algorithm is
associated with permanent ground deformation (PGD). Note that in both of these
algorithms the diameter of pipe is not considered to be a factor.

The PGV algorithm is based on the empirical data presented in a work done by
O'Rourke and Ayala (1993). The data correspond to actual pipeline damage
observed in four US and two Mexican earthquakes. This data is plotted in Figure
8.22.a. The following relation represents a good fit for this empircal data:

Repair Rate [Repairs/Km] 0.0001 x (PGV)
(2.25)


With PGV expressed in cm/sec. Note that the data plotted in Figure 8.22.a
correspond to asbestos cement, concrete and cast iron pipes; therefore, the above
(RR to PGV) relation is assumed to apply for brittle pipelines. For ductile
pipelines (steel, ductile iron and PVC), the above relation is multiplied by 0.3.
That is, ductile pipelines have 30% of the vulnerability of brittle pipelines. Note
that welded steel pipes with arc-welded joints are classified as ductile, and that
welded steel pipes with gas-welded joints are classified as brittle. It is
conceivable that the only other information available to the user regarding steel
pipes is the year of installation. In this case, the user should classify pre-1935
steel pipes as brittle pipes.

The damage algorithm for buried pipelines due to ground failure is based on work
conducted by Honegger and Eguchi (1992) for the San Diego County Water
Authority (SDCWA). Figure 8.22.b shows the base fragility curve for cast iron
pipes. The best-fit function to this curve is given by:

Repair Rate [Repairs/Km] Prob [liq] x PGD
(0.56)


With PGD expressed in inches. This RR to PGD relation is assumed to apply for
brittle pipelines. For ductile pipelines, the same multiplier as the PGV algorithm
is assumed (i.e., 0.3).

To summarize, the pipeline damage algorithms that are used in the current loss
estimation methodology are presented in Table 8.10

818
Chapter8DirectDamagetoLifelinesUtilitySystems
Table 8.10: Damage Algorithms for Water Pipelines

PGV Algorithm PGD Algorithm

R. R. 0.0001 x PGV
(2.25)

R. R. Prob[liq]xPGD
(0.56)

Pipe Type
Multiplier
Example of
Pipe
Multiplier
Example of
Pipe
Brittle Pipes
(PWP1)
1 CI, AC, RCC 1 CI, AC, RCC
Ductile Pipes
(PWP2)
0.3 DI, S, PVC 0.3 DI, S, PVC

8.1.9 System Performance

In the previous section, damage algorithms for the various components of a water system
were presented. For the level 2 enhanced analysis (i.e., assuming the commercial data
was readily available and processed as described in the Potable Water System Analysis
Model manual), this information is combined and a system network analysis is
performed.

This section, however, outlines the simplified methodology that is used in the level 1 and
level 2 analyses and which allows for a quick evaluation of the system performance in the
aftermath of an earthquake.

This approach is based on system performance studies done for water networks in
Oakland, Tokyo, and San Francisco. In the Tokyo study (Isoyama and Katayama, 1982),
water system network performance evaluations following an earthquake were simulated
for two different supply strategies: (1) supply priority to nodes with larger demands, and
(2) supply priority to nodes with lowest demands. The "best" and "worst" node
performances are approximately reproduced in a different format in Figure 8.23. The
probability of pipeline failure, which was assumed to follow a Poisson process in the
original paper, was substituded with the average break rate which was backcalculated
based on a pipeline link length of about 5 kilometers (i.e., in the trunk network of the
water supply system of Tokyo, the average link length is about 5 kilometers). Note that
in this figure, serviceability index is considered as a measure of the reduced flow.

Recently, researchers at Cornell University (Markov, Grigoriu and O'Rourke, 1994)
evaluated the San Francisco auxiliary (fire fighting) water supply system (AWSS). Some
of their results are reproduced and shown also in Figure 8.23.

G&E (1994) also did a similar study for the EBMUD (East Bay Municipal District) water
supply system. Their results are shown as well in Figure 8.23.

Based on these results, the damage algorithm proposed in this earthquake loss estimation
for the simplified system performance evaluation is defined by a "conjugate" lognormal
function (i.e., 1 - lognormal function). This damage function has a median of 0.1
repairs/km and a beta of 0.85, and it is shown in Figure 8.23. Hence, given knowledge of
819
HazusMHTechnicalManual
the pipe classification and length, one can estimate the system performance. That is,
damage algorithms provided in the previous section give repair rates and therefore the
expected total number of repairs (i.e., by multiplying the expected repair rate for each
pipe type in the network by its length and summing up over all pipes in the network).
The average repair rate is then computed as the ratio of the expected total number of
repairs to the total length of pipes in the network.

Example
Assume we have a pipeline network of total length equal to 500 kilometers, and that this
network is mainly composed of 16" diameter brittle pipes with each segment being 20
feet in length. Assume also that this pipeline is subject to both ground shaking and
ground failure as detailed in Table 8.11. Note that the repair rates (R.R.) in this table are
computed based on the equations provided in section 8.1.8.

Table 8.11: Example of System Performance Evaluation
PGV
(cm/sec)
R.R.
(Re/km)
Length
(km)
#
Repairs
PGD
(inches)
Probab.
of Lique
R.R.
(Re/km)
Length
(km)
#
Repairs
35 0.2980 50 ~15 18 1.0 5.0461 1 ~5
30 0.2106 50 ~11 12 1.0 4.0211 1 ~4
25 0.1398 50 ~7 6 0.80 2.7275 5 ~11
20 0.0846 50 ~4 2 0.65 1.4743 53 ~51
15 0.0443 100 ~4 1 0.60 1 20 12
10 0.0178 100 ~2 0.5 0.40 0.6783 20 ~6
5 0.0038 100 0 0 0.10 0 400 0
Total 500 43 Total 500 89

Therefore, due to PGV, the estimated number of leaks is 80% x 43 =34, and the
estimated number of breaks is 9, while due to PGD, the estimated number of leaks is 20%
x 89 =18 and the estimated number of breaks is 71.

When we apply the "conjugate" lognormal damage function, which has a median of 0.1
repairs/km and a beta of 0.85, first we compute conservatively the average break rate as:

Average break rate =(9 +71) / 500 =0.16 repairs/km

Hence, the serviceability index right after the earthquake is:

Serviceability Index =1 - Lognormal(0.16, 0.1, 0.85) =0.29 or 29 %

8.1.10 Guidance for Loss Estimation Using Advanced Data and Models Analysis

For this type of analysis, the expert can use the methodology developed with the
flexibility to (1) include a more refined inventory of the water system pertaining to the
area of study, (2) include component-specific and system-specific fragility data, and (3)
utilize a commercial model to estimate overall system functionality. Default damage
algorithms can be modified or replaced to incorporate improved information about key
components of a water system. Similarly, better restoration curves can be developed,
820
Chapter8DirectDamagetoLifelinesUtilitySystems
given knowledge of available resources and a more accurate layout of the water network
within the local topographic and geological conditions.

8.1.11 References

(1) ATC-13, "Earthquake Damage Evaluation Data for California", Applied Technology Council,
Redwood City, CA, 1985.

(2) G&E Engineering Systems, Inc. (G&E), "NIBS Earthquake Loss Estimation Methods,
Technical Manual, (Water Systems)", May 1994.

(3) Eguchi R.T., Taylor C., and Hasselman T.K., "Seismic Component Vulnerability Models for
Lifeline Risk Analysis", February 1983.

(4) Velon J .P. and J ohnson T.J ., "Water Distribution and Treatment (Davis Handbook of Applied
Hydraulics)", 1993.

(5) Honegger D.G. and Eguchi R.T., "Determination of Relative Vulnerabilities to Seismic
Damage for San Diego County Water Authority (SDCWA) Water Transmission Pipelines",
October 1992.

(6) McGhee T.J ., " Water Supply and Sewerage, 6th Edition", 1993.

(7) K/J /C and Dames & Moore, "Earthquake Loss Estimation for the City of Everett, Washington
Lifelines", May 1991.

(8) K/J /C and Dames & Moore, "Earthquake Loss Estimation Modeling of the Seattle Water
System", October 1990.

(9) TCLEE, "Seismic Loss Estimates for a Hypothetical Water System", August 1991.

(10) O'Rourke M.J and Ayala G., "Pipeline Damage due to Wave Propagation J ournal of
Geotechnical Engineering, ASCE Vol 119, No.9, Sept. 1993.

(11) Markov I., Grigoriu M., and O'Rourke T., "An evaluation of Seismic Serviceability of Water
Supply Networks with Application to San Francisco Auxiliary Water Supply System", NCEER
Report No. 94-0001, 1994.

(12) Isoyama R. and Katayama T., "Reliability Evaluation of Water Supply Systems During
Earthquakes", February 1982.

(13) Khater M and Waisman F., "Potable Water System Analysis Model (POWSAM)",
September 1999. NIBS technical report.


821
HazusMHTechnicalManual
Time (days)

P
e
r
c
e
n
t

F
u
n
c
t
i
o
n
a
l

0
25
50
75
100
1 10 100 1000
Minor Moderate Extensive Complete ATC-13 Data

Figure 8.1: Restoration Curves for Water Treatment Plants (after ATC-13, 1985).

Time (days)

P
e
r
c
e
n
t

F
u
n
c
t
i
o
n
a
l

0
25
50
75
100
1 10 100 1000
Minor Moderate Extensive Complete ATC-13 Data

Figure 8.2: Restoration Curves for Pumping Plants (after ATC-13, 1985).

822
Chapter8DirectDamagetoLifelinesUtilitySystems
Time (days)

P
e
r
c
e
n
t

F
u
n
c
t
i
o
n
a
l

0
25
50
75
100
1 10 100 1000
Minor Moderate Extensive Complete ATC-13 Data

Figure 8.3: Restoration Curves for Wells (after ATC-13, 1985).


Time (days)

P
e
r
c
e
n
t

F
u
n
c
t
i
o
n
a
l

0
25
50
75
100
1 10 100 1000
Minor Moderate Extensive Complete ATC-13 Data

Figure 8.4: Restoration Curves for Water Storage Tanks (after ATC-13, 1985).






823
HazusMHTechnicalManual
Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y


D
s

>

d
s

|


P
G
A

]



0.000
0.250
0.500
0.750
1.000
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive Complete

Figure 8.5: Fragility Curves for Small Water Treatment Plants with Anchored
Components.


Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y


D
s

>

d
s

|


P
G
A

]



0.000
0.250
0.500
0.750
1.000
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive Complete

Figure 8.6: Fragility Curves for Small Water Treatment Plants with Unanchored
Components.


824
Chapter8DirectDamagetoLifelinesUtilitySystems
Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y


D
s

>

d
s

|


P
G
A

]



0.000
0.250
0.500
0.750
1.000
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive Complete

Figure 8.7: Fragility Curves for Medium Water Treatment Plants with Anchored
Components.


Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y


D
s

>

d
s

|


P
G
A

]



0.000
0.250
0.500
0.750
1.000
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive Complete

Figure 8.8: Fragility Curves for Medium Water Treatment Plants with Unanchored
Components.
825
HazusMHTechnicalManual
Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y


D
s

>

d
s

|


P
G
A

]



0.000
0.250
0.500
0.750
1.000
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive Complete

Figure 8.9: Fragility Curves for Large Water Treatment Plants with Anchored
Components.


Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y


D
s

>

d
s

|


P
G
A

]



0.000
0.250
0.500
0.750
1.000
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive Complete

Figure 8.10: Fragility Curves for Large Water Treatment Plants with Unanchored
Components.



826
Chapter8DirectDamagetoLifelinesUtilitySystems
Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y


D
s

>

d
s

|


P
G
A

]



0.000
0.250
0.500
0.750
1.000
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive Complete

Figure 8.11: Fragility Curves for Small Pumping Plants with Anchored
Components.



Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y


D
s

>

d
s

|


P
G
A

]



0.000
0.250
0.500
0.750
1.000
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive Complete

Figure 8.12: Fragility Curves for Small Pumping Plants with Unanchored
Components.



827
HazusMHTechnicalManual
Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y


D
s

>

d
s

|


P
G
A

]



0.000
0.250
0.500
0.750
1.000
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive Complete

Figure 8.13: Fragility Curves for Medium/Large Pumping Plants with Anchored
Components.

Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y


D
s

>

d
s

|


P
G
A

]



0.000
0.250
0.500
0.750
1.000
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive Complete

Figure 8.14: Fragility Curves for Medium/Large Pumping Plants with Anchored
Components.





828
Chapter8DirectDamagetoLifelinesUtilitySystems

Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y


D
s

>

d
s

|


P
G
A

]



0.000
0.250
0.500
0.750
1.000
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive Complete


Figure 8.15: Fragility Curves for Wells


Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y


D
s

>

d
s

|


P
G
A

]



0.000
0.250
0.500
0.750
1.000
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive Complete


Figure 8.16: Fragility Curves for Anchored On Ground Concrete Tank.

829
HazusMHTechnicalManual
Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y


D
s

>

d
s

|


P
G
A

]



0.000
0.250
0.500
0.750
1.000
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive Complete

Figure 8.17: Fragility Curves for Unanchored On Ground Concrete Tank.



Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y


D
s

>

d
s

|


P
G
A

]



0.000
0.250
0.500
0.750
1.000
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive Complete


Figure 8.18: Fragility Curves for Anchored On Ground Steel Tank.


830
Chapter8DirectDamagetoLifelinesUtilitySystems
Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y


D
s

>

d
s

|


P
G
A

]



0.000
0.250
0.500
0.750
1.000
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive Complete

Figure 8.19: Fragility Curves for Unanchored On Ground Steel Tank.


Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y


D
s

>

d
s

|


P
G
A

]



0.000
0.250
0.500
0.750
1.000
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive Complete

Figure 8.20: Fragility Curves for Above Ground Steel Tank.

831
HazusMHTechnicalManual
Peak Ground Acceleration (g)
[

P
r
o
b
a
b
i
l
i
t
y


D
s

>

d
s

|


P
G
A

]



0.000
0.250
0.500
0.750
1.000
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Minor Moderate Extensive Complete

Figure 8.21: Fragility Curves for On Ground Wood Tank.




832
Chapter8DirectDamagetoLifelinesUtilitySystems

Peak Ground Velocity (cm/sec)
R
e
p
a
i
r

R
a
t
e

[

R
e
p
a
i
r
s
/
K
m

]





0.0001
0.001
0.01
0.1
1
10
1 10 100

Figure 8.22.a: Ground Shaking (Wave Propagation) Damage Model for Brittle Pipes
(Specifically CI, AC, RCC, and PCCP) Based on Four U.S. and Two Mexican
Earthquakes (after O'Rourke and Ayala, 1993).

Permanent Ground Deformation (inches)
R
e
p
a
i
r

R
a
t
e

[
R
e
p
a
i
r
s
/
k
m
]




0
4
8
12
16
0 20 40 60 80 100
Honegger & Eguchi (1992)
NIBS

Figure 8.22.b: Ground Deformation Damage Model for Cast Iron Pipes (after
Honegger and Eguchi, 1992).


833
HazusMHTechnicalManual








Average Break Rate [Breaks/km]
S
e
r
v
i
c
e
a
b
i
l
i
t
y

I
n
d
e
x

[
P
e
r
c
e
n
t
a
g
e
]






0
20
40
60
80
100
0.01 0.1 1
NIBS
AWSS (simulation results at
Cornell)
AWSS (average)
EBMUD(G&E) [Large Pipes]
EBMUD (G&E) [Small Pipes]
Isoyama & Katayama (lower
range)
Isoyama & Katayama (upper
range)

Figure 8.23: Damage Index Versus Ave